File size: 1,364 Bytes
865b874 69c25a8 865b874 69c25a8 865b874 69c25a8 234c0fa 69c25a8 234c0fa 69c25a8 234c0fa 69c25a8 234c0fa |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 |
---
language:
- en
- ru
license: apache-2.0
tags:
- gpt
- NLG
---
# HuYaLM 100B
**Hu**gging Face **YaLM 100B** (by [BlackSamorez](https://github.com/BlackSamorez)) is a **transformers-compatible** implementation of the **YaLM 100B** model. Originally trained by Yandex, the model used 800 A100 graphics cards and 1.7 TB of diverse text data, including online texts and books, in both English and Russian.
The motivation behind this particular implementation is to update the originally published, outdated code to align with the latest advancements in the field. As this code is compatible with the _transformers_ library, it inherently supports crucial features like [quantization](https://huggingface.co/docs/transformers/main_classes/quantization) (for model size optimization) and [adapter training](https://huggingface.co/docs/peft/index) (for efficient fine-tuning).
For more details on training, acceleration, and stabilization techniques, you can refer to articles on **[Medium](https://medium.com/p/d1df53d0e9a6)** (in English) and **[Habr](https://habr.com/ru/company/yandex/blog/672396/)** (in Russian). The original code from Yandex is available on [GitHub](https://github.com/yandex/YaLM-100B).
This code and model are distributed under the [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0) license, which allows for commercial use. |