meditron-70b-chat / README.md
malhajar's picture
Update README.md
481da13 verified
|
raw
history blame
No virus
1.81 kB
---
language:
- en
tags:
- Medicine
datasets:
- malhajar/alpaca-gpt4-tr
license: llama2
base_model: epfl-llm/meditron-70b
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
meditron-7b-chat is a finetuned version of [`epfl-llm/meditron-70b`](https://huggingface.co/epfl-llm/meditron-70b) using SFT Training on the Alpaca Dataset.
This model can answer information about different excplicit ideas in medicine (see [`epfl-llm/meditron-70b`](https://huggingface.co/epfl-llm/meditron-70b) for more info)
### Model Description
- **Finetuned by:** [`Mohamad Alhajar`](https://www.linkedin.com/in/muhammet-alhajar/)
- **Language(s) (NLP):** English
- **Finetuned from model:** [`epfl-llm/meditron-70b`](https://huggingface.co/epfl-llm/meditron-70b)
### Prompt Template
```
### Instruction:
<prompt> (without the <>)
### Response:
```
## How to Get Started with the Model
Use the code sample provided in the original post to interact with the model.
```python
from transformers import AutoTokenizer,AutoModelForCausalLM
model_id = "malhajar/meditron-70b-chat"
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
device_map="auto",
torch_dtype=torch.float16,
revision="main")
tokenizer = AutoTokenizer.from_pretrained(model_id)
question: "what is tract infection?"
# For generating a response
prompt = '''
### Instruction:
{question}
### Response:'''
input_ids = tokenizer(prompt, return_tensors="pt").input_ids
output = model.generate(inputs=input_ids,max_new_tokens=512,pad_token_id=tokenizer.eos_token_id,top_k=50, do_sample=True,
top_p=0.95)
response = tokenizer.decode(output[0])
print(response)
```