llama-2-7b-chat-MEDS-12

This is a llama-2-7b-chat-hf model fine-tuned using QLoRA (4-bit precision) on the s200862/medical_qa_meds dataset. This is an adapted version of the medalpaca/medical_meadow_wikidoc_patient_information dataset to match llama-2's instruction format.

🔧 Training

It was trained on-premise in a jupyter notebook using an Nvidia RTX A4000 GPU with 16GB of VRAM and 16 GB of system RAM.

💻 Usage

It is intended to give answers to medical questions.

# pip install transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "s200862/llama-2-7b-chat-MEDS-12"
prompt = "What causes Allergy?"

tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

sequences = pipeline(
    f'<s>[INST] {prompt} [/INST]',
    do_sample=True,
    top_k=10,
    num_return_sequences=1,
    eos_token_id=tokenizer.eos_token_id,
    max_length=200,
)
for seq in sequences:
    print(f"Result: {seq['generated_text']}")