codeparrot/code-generation-models

1[PolyCoder](https://github.com/VHellendoorn/Code-LMs) uses GPT2 architecture, with BPE tokenizer trained on a random 5% subset of the data (all languages), and a context length of 2048. To study the effect of scaling of model size, the odel was trained in 3 different sizes.2 3<div align="center">4 5|Model | # parameters | 6|   -   |   -  | 7| GPT2 | 160M |8| GPT2 | 400M |9| GPT2 | 2.7B |10 11</div>12 13 14PolyCoder is currently being integrated in 🤗 `transformers`. Meanwhile it can be loaded following the instructions in the original GitHub [repo](https://github.com/vhellendoorn/code-lms#models).