RichardErkhov/starmpcc_-_Asclepius-Llama3-8B-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Asclepius-Llama3-8B - GGUF

• Model creator: https://huggingface.co/starmpcc/
• Original model: https://huggingface.co/starmpcc/Asclepius-Llama3-8B/

Name	Quant method	Size
Asclepius-Llama3-8B.Q2_K.gguf	Q2_K	2.96GB
Asclepius-Llama3-8B.IQ3_XS.gguf	IQ3_XS	3.28GB
Asclepius-Llama3-8B.IQ3_S.gguf	IQ3_S	3.43GB
Asclepius-Llama3-8B.Q3_K_S.gguf	Q3_K_S	3.41GB
Asclepius-Llama3-8B.IQ3_M.gguf	IQ3_M	3.52GB
Asclepius-Llama3-8B.Q3_K.gguf	Q3_K	3.74GB
Asclepius-Llama3-8B.Q3_K_M.gguf	Q3_K_M	3.74GB
Asclepius-Llama3-8B.Q3_K_L.gguf	Q3_K_L	4.03GB
Asclepius-Llama3-8B.IQ4_XS.gguf	IQ4_XS	4.18GB
Asclepius-Llama3-8B.Q4_0.gguf	Q4_0	4.34GB
Asclepius-Llama3-8B.IQ4_NL.gguf	IQ4_NL	4.38GB
Asclepius-Llama3-8B.Q4_K_S.gguf	Q4_K_S	4.37GB
Asclepius-Llama3-8B.Q4_K.gguf	Q4_K	4.58GB
Asclepius-Llama3-8B.Q4_K_M.gguf	Q4_K_M	4.58GB
Asclepius-Llama3-8B.Q4_1.gguf	Q4_1	4.78GB
Asclepius-Llama3-8B.Q5_0.gguf	Q5_0	5.21GB
Asclepius-Llama3-8B.Q5_K_S.gguf	Q5_K_S	5.21GB
Asclepius-Llama3-8B.Q5_K.gguf	Q5_K	5.34GB
Asclepius-Llama3-8B.Q5_K_M.gguf	Q5_K_M	5.34GB
Asclepius-Llama3-8B.Q5_1.gguf	Q5_1	5.65GB
Asclepius-Llama3-8B.Q6_K.gguf	Q6_K	6.14GB
Asclepius-Llama3-8B.Q8_0.gguf	Q8_0	7.95GB

Original model description:

license: cc-by-nc-sa-4.0 datasets: - starmpcc/Asclepius-Synthetic-Clinical-Notes language: - en pipeline_tag: text2text-generation tags: - medical

Model Card for Model ID

This is an official model checkpoint for Asclepius-Llama3-8B (arxiv). This model is an enhanced version of Asclepius-7B, by replacing the base model with Llama-3 and increasing the max sequence length to 8192.

UPDATE

2024.01.10

• Asclepius-R, the variant of Asclepius that trained on MIMIC-III discharge summaries, is now available on Physionet!

Model Details

Model Description

• Model type: Clinical LLM (Large Language Model)
• Language(s) (NLP): English
• License: CC-BY-NC-SA 4.0
• Finetuned from model [optional]: Llama3-8B

Model Sources [optional]

• Repository: https://github.com/starmpcc/Asclepius
• Paper: https://arxiv.org/abs/2309.00237
• Data: https://huggingface.co/datasets/starmpcc/Asclepius-Synthetic-Clinical-Notes

Uses

This model can perform below 8 clinical NLP tasks, with clincal notes.

• Named Entity Recognition
• Abbreviation Expansion
• Relation Extraction
• Temporal Information Extraction
• Coreference Resolution
• Paraphrasing
• Summarization
• Question Answering

Direct Use

[More Information Needed]

Downstream Use [optional]

[More Information Needed]

Out-of-Scope Use

ONLY USE THIS MODEL FOR RESEARCH PURPOSE!!

How to Get Started with the Model

prompt = """You are an intelligent clinical languge model.
Below is a snippet of patient's discharge summary and a following instruction from healthcare professional.
Write a response that appropriately completes the instruction.
The response should provide the accurate answer to the instruction, while being concise.

[Discharge Summary Begin]
{note}
[Discharge Summary End]

[Instruction Begin]
{question}
[Instruction End] 
"""

from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("starmpcc/Asclepius-Llama3-8B", use_fast=False)
model = AutoModelForCausalLM.from_pretrained("starmpcc/Asclepius-Llama3-8B")

note = "This is a sample note"
question = "What is the diagnosis?"

model_input = prompt.format(note=note, question=question)
input_ids = tokenizer(model_input, return_tensors="pt").input_ids
output = model.generate(input_ids)
print(tokenizer.decode(output[0]))

Training Details

Training Data

https://huggingface.co/datasets/starmpcc/Asclepius-Synthetic-Clinical-Notes

Training Procedure

• Initial training was conducted using causal language modeling on synthetic clinical notes.
• It was then fine-tuned with clinical instruction-response pairs.
• For a comprehensive overview of our methods, our upcoming paper will serve as a resource.

Training Hyperparameters

• We followed config used in Stanford Alpaca

Speeds, Sizes, Times

• Pre-Training (1 epoch): 2h 59m with 4x A100 80G
• Instruction Fine-Tuning (3 epoch): 30h 41m with 4x A100 80G

Citation

BibTeX:

@article{kweon2023publicly,
  title={Publicly Shareable Clinical Large Language Model Built on Synthetic Clinical Notes},
  author={Kweon, Sunjun and Kim, Junu and Kim, Jiyoun and Im, Sujeong and Cho, Eunbyeol and Bae, Seongsu and Oh, Jungwoo and Lee, Gyubok and Moon, Jong Hak and You, Seng Chan and others},
  journal={arXiv preprint arXiv:2309.00237},
  year={2023}
}