ZiweiChen/BioMistral-Clinical-7B

Model Card for Model ID

How to use

Loading the model from Hunggingface:

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ZiweiChen/BioMistral-Clinical-7B")
model = AutoModelForCausalLM.from_pretrained("ZiweiChen/BioMistral-Clinical-7B")

Lightweight model loading can be used - using 4-bit quantization!

!pip install -q -U bitsandbytes
!pip install -q -U git+https://github.com/huggingface/transformers.git
!pip install -q -U git+https://github.com/huggingface/peft.git
!pip install -q -U git+https://github.com/huggingface/accelerate.git

from transformers import  AutoTokenizer, BitsAndBytesConfig, AutoModelForCausalLM
import torch

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_use_double_quant=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16
)

tokenizer = AutoTokenizer.from_pretrained("ZiweiChen/BioMistral-Clinical-7B")
model = AutoModelForCausalLM.from_pretrained("ZiweiChen/BioMistral-Clinical-7B", quantization_config=bnb_config)

How to Generate text:

model_device = next(model.parameters()).device

prompt = """
### Question:

How to treat severe obesity?

### Answer:
"""
model_input = tokenizer(prompt, return_tensors="pt").to(model_device)

with torch.no_grad():
    output = model.generate(**model_input, max_new_tokens=100)
    answer = tokenizer.decode(output[0], skip_special_tokens=True)
    print(answer)

Model Details

Model Description