|
--- |
|
license: apache-2.0 |
|
--- |
|
|
|
|
|
|
|
|
|
|
|
APUS-xDAN-4.0-MOE |
|
Introduction |
|
APUS-xDAN-4.0-MOE is a transformer-based MoE decoder-only language model pretrained on a large amount of data. |
|
|
|
For more details, please refer to our blog post and GitHub repo. |
|
|
|
Model Details |
|
APUS-xDAN-4.0-MOE employs Mixture of Experts (MoE) architecture, where the models are upcycled from dense language models. For instance, APUS-xDAN-4.0-MOE is upcycled from Qwen-1.8B. It has 14.3B parameters in total and 2.7B activated parameters during runtime, while achieching comparable performance to Qwen1.5-7B, it only requires 25% of the training resources. We also observed that the inference speed is 1.74 times that of Qwen1.5-7B. |
|
|
|
Requirements |
|
The code of APUS-xDAN-4.0-MOE has been in the latest Hugging face transformers and we advise you to build from source with command pip install git+https://github.com/huggingface/transformers, or you might encounter the following error: |
|
|
|
|
|
|
|
Usage |
|
We do not advise you to use base language models for text generation. Instead, you can apply post-training, e.g., SFT, RLHF, continued pretraining, etc., on this model. |
