LLaMA 3.2 3B Instruct - Healthcare Fine-tuned Model
This is a model that fine-tuned the Llama-3.2-3B-Instruct model from Unidocs using Healthcare data.
μ λλ₯μ€(μ£Ό)μμ Llama-3.2-3B-Instruct λͺ¨λΈμ Healthcare λ°μ΄ν°λ‘ λ―ΈμΈμ‘°μ ν λͺ¨λΈμ
Model Description
sLLM model used in Unidoc's ezMyAIDoctor, released on December 10, 2024 as a result of the AIDC-HPC project
of the Artificial Intelligence Industry Convergence Business Group (AICA)
meta-llama/Llama-3.2-3B-Instruct wiki, kowiki, super-large AI healthcare question-answer data,
A model that has been pretrained (Full Finetuning) by referring to the super-large AI corpus with improved Korean performance,
and the medical and legal professional book corpus.
μ λλ₯μ€(μ£Ό)μ ezMyAIDoctorμμ μ¬μ©λλ sLLM λͺ¨λΈλ‘ μΈκ³΅μ§λ₯μ°μ
μ΅ν©μ¬μ
λ¨(AICA)μ AIDC-HPC μ¬μ
μ κ²°κ³Όλ‘ 2024λ
12μ 10μΌ κ³΅κ°ν¨
meta-llama/Llama-3.2-3B-Instructμ wiki, kowiki, AIHub(aihub.or.kr)μ (μ΄κ±°λAI ν¬μ€μΌμ΄ μ§μμλ΅λ°μ΄ν°,
νκ΅μ΄ μ±λ₯μ΄ κ°μ λ μ΄κ±°λ AI λ§λμΉ, μλ£/λ²λ₯ μ λ¬Έμμ λ§λμΉ)λ₯Ό μ°Έκ³ νμ¬ Pretrain(Full Finetuning)λ λͺ¨λΈμ
Intended Uses & Limitations
The model is designed to assist with healthcare-related queries and tasks.
However, it should not be used as a substitute for professional medical advice, diagnosis, or treatment.
Always consult with a qualified healthcare provider for medical concerns.
μ΄ λͺ¨λΈμ Healthcare κ΄λ ¨ μ§μ λ° μμ
μ μ§μνλλ‘ μ€κ³λμμ΅λλ€.
κ·Έλ¬λ μ λ¬Έμ μΈ μνμ μ‘°μΈ, μ§λ¨ λλ μΉλ£λ₯Ό λ체νλ λ° μ¬μ©λμ΄μλ μ λ©λλ€.
μλ£ κ΄λ ¨ λ¬Έμ λ νμ μ격μ κ°μΆ μλ£ μλΉμ€ μ 곡μμ μμνμμμ€.
Training Data
The model was fine-tuned on a proprietary healthcare dataset.
Due to privacy concerns, details of the dataset cannot be disclosed.
wiki, kowiki λ°μ΄ν° μ΄μΈ
κ³ΌνκΈ°μ μ 보ν΅μ λΆ, νκ΅μ§λ₯μ 보μ¬νμ§ν₯μμμ κ΄λ¦¬νκ³ μλ AIHubμ
- μ΄κ±°λAI ν¬μ€μΌμ΄ μ§μμλ΅λ°μ΄ν°
- νκ΅μ΄ μ±λ₯μ΄ κ°μ λ μ΄κ±°λ AI λ§λμΉ
- μλ£, λ²λ₯ μ λ¬Έμμ λ§λμΉ
λ±μ νμ©ν¨
Training Procedure
Full fine-tuning was performed on the base LLaMA 3.2 3B Instruct model using the healthcare dataset.
Healthcare λ°μ΄ν° μΈνΈλ₯Ό μ¬μ©νμ¬ κΈ°λ³Έ LLaMA 3.2 3B Instruct λͺ¨λΈμμ μ 체 λ―ΈμΈ μ‘°μ μ μννμ΅λλ€.
Evaluation Results
Accuracy by category of mmlu benchmark
category | Accuracy |
---|---|
anatomy | 0.54 (73/135) |
clinical_knowledge | 0.59 (156/265) |
college_medicine | 0.57 (99/173) |
medical_genetics | 0.64 (64/100) |
professional_medicine | 0.60 (162/272) |
All Accuracy Mean value: 0.59
Use with transformers
Starting with transformers >= 4.43.1
onward, you can run conversational inference using the Transformers pipeline
abstraction or by leveraging the Auto classes with the generate()
function.
Make sure to update your transformers installation via pip install --upgrade transformers
.
import transformers
import torch
model_id = "unidocs/llama-3.2-3b-komedic-instruct"
pipeline = transformers.pipeline(
"text-generation",
model=model_id,
model_kwargs={"torch_dtype": torch.bfloat16},
device_map="auto",
)
messages = [
{"role": "system", "content": "λΉμ μ μλ£μ λ¬Έκ°μ
λλ€. μ§λ³μ μ μ, μμΈ, μ¦μ, κ²μ§, μ§λ¨, μΉλ£, μ½λ¬Ό, μμ΄, μν μΈ‘λ©΄μμ λ΅λ³ν΄ μ£ΌμΈμ"},
{"role": "user", "content": "곡볡νλΉμ΄ 120μ΄μμΈ κ²½μ° μ 1ν λΉλ¨μ μ 2ν λΉλ¨ νμλ κ°κ° μ΄λ»κ² μΉλ£λ₯Ό λ°μμΌ νλμ?"},
]
outputs = pipeline(
messages,
max_new_tokens=256,
)
print(outputs[0]["generated_text"][-1])
Note: You can also find detailed recipes on how to use the model locally, with torch.compile()
, assisted generations, quantised and more at huggingface-llama-recipes
Limitations and Bias
This model may produce biased or inaccurate results. It should not be solely relied upon for critical healthcare decisions.
The model's knowledge is limited to its training data and cut-off date.
It may exhibit biases present in the training data.
The model may occasionally produce incorrect or inconsistent information.
λͺ¨λΈμ μ§μμ νλ ¨ λ°μ΄ν°μ λ§κ°μΌλ‘ μ νλ©λλ€.
νλ ¨ λ°μ΄ν°μ νΈν₯μ΄ μμ μ μμ΅λλ€.
λͺ¨λΈμ κ°λ μλͺ»λκ±°λ μΌκ΄λμ§ μμ μ 보λ₯Ό μμ±ν μ μμ΅λλ€.
μ΄ λͺ¨λΈμ νΈν₯λκ±°λ λΆμ νν κ²°κ³Όλ₯Ό μμ±ν μ μμ΅λλ€. μ€μν μλ£ κ²°μ μ μ΄ λͺ¨λΈμλ§ μμ‘΄ν΄μλ μ λ©λλ€.
Legal Disclaimer
The model developers and distributors bear no legal responsibility for any consequences arising from the use of this model.
This includes any direct, indirect, incidental, special, punitive, or consequential damages resulting from the model's output.
By using this model, users assume all risks that may arise, and the responsibility for verifying and appropriately using the model's output lies solely with the user.
This model cannot substitute for medical advice, diagnosis, or treatment, and qualified healthcare professionals should always be consulted for medical decisions.
This disclaimer applies to the maximum extent permitted by applicable law.
λ²μ μ± μ λ©΄μ± μ‘°ν
λ³Έ λͺ¨λΈμ μ¬μ©μΌλ‘ μΈν΄ λ°μνλ λͺ¨λ κ²°κ³Όμ λν΄ λͺ¨λΈ κ°λ°μ λ° λ°°ν¬μλ μ΄λ ν λ²μ μ±
μλ μ§μ§ μμ΅λλ€.
μ΄λ λͺ¨λΈμ μΆλ ₯μΌλ‘ μΈν μ§μ μ , κ°μ μ , μ°λ°μ , νΉμν, μ§λ²μ λλ κ²°κ³Όμ μν΄λ₯Ό ν¬ν¨ν©λλ€.
μ¬μ©μλ λ³Έ λͺ¨λΈμ μ¬μ©ν¨μΌλ‘μ¨ λ°μν μ μλ λͺ¨λ μνμ κ°μνλ©°, λͺ¨λΈμ μΆλ ₯μ λν κ²μ¦ λ° μ μ ν μ¬μ©μ λν μ±
μμ μ μ μΌλ‘ μ¬μ©μμκ² μμ΅λλ€.
λ³Έ λͺ¨λΈμ μνμ μ‘°μΈ, μ§λ¨, λλ μΉλ£λ₯Ό λ체ν μ μμΌλ©°, μλ£ κ΄λ ¨ κ²°μ μ λ΄λ¦΄ λλ λ°λμ μ격μ κ°μΆ μλ£ μ λ¬Έκ°μ μλ΄ν΄μΌ ν©λλ€.
μ΄ λ©΄μ±
μ‘°νμ κ΄λ ¨ λ²λ₯ μ΄ νμ©νλ μ΅λ λ²μ λ΄μμ μ μ©λ©λλ€.
Model Card Contact
μ μ ([email protected]), κΉμ§μ€([email protected]),
κΉμ’
μ([email protected])
Additional Information
For more details about the base model, please refer to the original LLaMA 3.2 documentation.
- Downloads last month
- 23