lonepatient / awesome-pretrained-chinese-nlp-models Goto Github PK

和Chinese-Transformer-XL师出同门的，评估下是否要收录
https://github.com/THUDM/GLM-130B

Thanks for your great work on collecting such lots of LLM and dataset!
Here is my suggestion: Since the dataset format of PT, SFT, RW and DPO/PPO stage is not the same, it will be clear if add a column "stage" to the dataset table.

https://github.com/THUDM/glm

https://github.com/scutcyr/BianQue
在ChatYuan的基础上微调的

https://bminf.readthedocs.io/zh_CN/latest/introduction-zh.html

信息你看看。适合收录不。

自荐：
https://github.com/BlinkDL/AI-Writer

Dear lonePatient,

I hope this message finds you well.

I want to express my gratitude for your valuable work. Your comprehensive survey has greatly contributed to the Chinese NLP community.

I am writing to request the inclusion of our open-source models, MAP-Neo and CT-LLM, in this GitHub repository.

Here is a brief introduction to our models:

MAP-Neo: A fully open-source model where all data for pre-training and post-training are open, along with the code and public model weights. More details are available on our project website MAP-Neo.
CT-LLM (Chinese Tiny LLM): The first pre-training model focused on Chinese corpus, also open-sourcing the Chinese pre-training dataset, data processing pipeline, intermediate checkpoints, and pre-training code. More information can be found on our website Chinese Tiny LLM.

Thank you for considering our request.

Best regards,

M-A-P Team