med_mistral_4bit

Model Card for med_mistral_4bit

<!-- Provide a quick summary of what the model is/does. -->

Model Details

Model Description

<!-- Provide a longer summary of what this model is. -->

Model 4-bit Mistral-7B-Instruct-v0.2 finetuned with QLoRA on multiple medical datasets.

16-bit version: med_mistral

• License: apache-2.0
• Finetuned from model : mistralai/Mistral-7B-Instruct-v0.2

Model Sources [optional]

<!-- Provide the basic links for the model. -->

• Repository: https://github.com/atadria/med_llm

Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

The model is finetuned on medical data and is intended only for research. It should not be used as a substitute for professional medical advice, diagnosis, or treatment.

Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

The model's predictions are based on the information available in the finetuned medical dataset. It may not generalize well to all medical conditions or diverse patient populations.

Sensitivity to variations in input data and potential biases present in the training data may impact the model's performance.

Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

How to Get Started with the Model

Use the code below to get started with the model.

# !pip install -q transformers accelerate bitsandbytes

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("adriata/med_mistral")
model = AutoModelForCausalLM.from_pretrained("adriata/med_mistral")

prompt_template = """<s>[INST] {prompt} [/INST]"""

prompt = "What is influenza?"

model_inputs = tokenizer.encode(prompt_template.format(prompt=prompt),
                                return_tensors="pt").to("cuda")

generated_ids = model.generate(model_inputs, max_new_tokens=512, do_sample=True)
decoded = tokenizer.batch_decode(generated_ids)
print(decoded[0])

Training Details

~13h - 20k examples x 1 epoch

GPU: OVH - 1 × NVIDIA TESLA V100S (32 GiB RAM)

Training Data

<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->

Training data included 20k examples randomly selected from datasets:

• pubmed
• bigbio/czi_drsm
• bigbio/bc5cdr
• bigbio/distemist
• pubmed_qa
• medmcqa