README.md

---
library_name: transformers
license: apache-2.0
base_model: PharMolix/BioMedGPT-LM-7B
language:
- en
---

16-bit version of weights from `PharMolix/BioMedGPT-LM-7B`, for easier download / finetuning / model-merging

Code

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

m2 = AutoModelForCausalLM.from_pretrained("PharMolix/BioMedGPT-LM-7B",
                                          torch_dtype=torch.float16,
                                          device_map="auto")
```