Hugging Face's logo

pfnet
/
Preferred-MedLLM-Qwen-72B

Text Generation
Transformers
Safetensors
English
Japanese
Model card Files
xet
Community
2
lbourdois commited on
Commit
b5d74db
·
verified ·
1 Parent(s): 0a334fe

Improve language tag

Browse files

Hi! As the model is multilingual, this is a PR to add other languages than English to the language tag to improve the referencing. Note that 29 languages are announced in the README, but only 13 are explicitly listed. I was therefore only able to add these 13 languages.

Files changed (1) hide show
  1. README.md +64 -53
README.md CHANGED
@@ -1,53 +1,64 @@
1
- ---
2
- base_model: Qwen/Qwen2.5-72B
3
- language:
4
- - en
5
- - ja
6
- library_name: transformers
7
- pipeline_tag: text-generation
8
- license: other
9
- license_name: qwen
10
- ---
11
-
12
- # Preferred-MedLLM-Qwen-72B
13
-
14
- ## Model Description
15
-
16
- Preferred-MedLLM-Qwen-72B is a finetuned model based on [Qwen/Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B), which has undergone continued pretraining on an original corpus of medical-related text.
17
-
18
- The model is released under the [Qwen LICENSE](https://huggingface.co/Qwen/Qwen2.5-72B/blob/main/LICENSE).
19
-
20
- ## Model Performance
21
-
22
- The table below shows the performance on the Japanese medical licensing examination from 2018 to 2022 ([IgakuQA](https://github.com/jungokasai/IgakuQA)).
23
- | Model ID | Average | 2018 | 2019 | 2020 | 2021 | 2022 |
24
- |:--------------------------------------------------------------------------------------------------------------------|-------------------:|-------:|-------:|-------:|-------:|-------:|
25
- | **Preferred-MedLLM-Qwen-72B** | **431.2** | **434** | 420 | **439** | **430** | 433 |
26
- | GPT-4o | 430.4 | 427 | **431** | 433 | 427 | **434** |
27
- | [Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B) | 398.4 | 412 | 394 | 394 | 393 | 399 |
28
- | [Llama3-Preferred-MedSwallow-70B](https://huggingface.co/pfnet/Llama3-Preferred-MedSwallow-70B) | 395.2 | 407 | 390 | 391 | 393 | 395 |
29
- | GPT-4 | 388.8 | 382 | 385 | 387 | 398 | 392 |
30
- | [Mistral-Large-Instruct-2407](https://huggingface.co/mistralai/Mistral-Large-Instruct-2407) | 376 | 370 | 371 | 390 | 373 | 376 |
31
- | [Llama-3.1-Swallow-70B-v0.1](https://huggingface.co/tokyotech-llm/Llama-3.1-Swallow-70B-v0.1) | 368.4 | 379 | 378 | 379 | 351 | 355 |
32
- | [Meta-Llama-3-70B](https://huggingface.co/meta-llama/Meta-Llama-3-70B) | 334.6 | 353 | 340 | 348 | 314 | 318 |
33
- | GPT-3.5 | 273.2 | 266 | 250 | 266 | 297 | 287 |
34
-
35
- ## Limitations
36
-
37
- The model was developed for research purposes and is not intended for clinical diagnosis.
38
- It is the users' responsibility to ensure compliance with applicable rules and regulations.
39
-
40
- ## Contributors
41
-
42
- Preferred Networks, Inc.
43
- - Junichiro Iwasawa
44
- - Wataru Kawakami
45
- - Keita Suzuki
46
-
47
- ## Publications
48
-
49
- We are currently preparing a related blog post and research paper, which will be made available soon to provide further insights and detailed information about the development and capabilities of this model.
50
-
51
- ## License
52
-
53
- [Qwen LICENSE](https://huggingface.co/Qwen/Qwen2.5-72B/blob/main/LICENSE)
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen2.5-72B
3
+ language:
4
+ - zho
5
+ - eng
6
+ - fra
7
+ - spa
8
+ - por
9
+ - deu
10
+ - ita
11
+ - rus
12
+ - jpn
13
+ - kor
14
+ - vie
15
+ - tha
16
+ - ara
17
+ library_name: transformers
18
+ pipeline_tag: text-generation
19
+ license: other
20
+ license_name: qwen
21
+ ---
22
+
23
+ # Preferred-MedLLM-Qwen-72B
24
+
25
+ ## Model Description
26
+
27
+ Preferred-MedLLM-Qwen-72B is a finetuned model based on [Qwen/Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B), which has undergone continued pretraining on an original corpus of medical-related text.
28
+
29
+ The model is released under the [Qwen LICENSE](https://huggingface.co/Qwen/Qwen2.5-72B/blob/main/LICENSE).
30
+
31
+ ## Model Performance
32
+
33
+ The table below shows the performance on the Japanese medical licensing examination from 2018 to 2022 ([IgakuQA](https://github.com/jungokasai/IgakuQA)).
34
+ | Model ID | Average | 2018 | 2019 | 2020 | 2021 | 2022 |
35
+ |:--------------------------------------------------------------------------------------------------------------------|-------------------:|-------:|-------:|-------:|-------:|-------:|
36
+ | **Preferred-MedLLM-Qwen-72B** | **431.2** | **434** | 420 | **439** | **430** | 433 |
37
+ | GPT-4o | 430.4 | 427 | **431** | 433 | 427 | **434** |
38
+ | [Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B) | 398.4 | 412 | 394 | 394 | 393 | 399 |
39
+ | [Llama3-Preferred-MedSwallow-70B](https://huggingface.co/pfnet/Llama3-Preferred-MedSwallow-70B) | 395.2 | 407 | 390 | 391 | 393 | 395 |
40
+ | GPT-4 | 388.8 | 382 | 385 | 387 | 398 | 392 |
41
+ | [Mistral-Large-Instruct-2407](https://huggingface.co/mistralai/Mistral-Large-Instruct-2407) | 376 | 370 | 371 | 390 | 373 | 376 |
42
+ | [Llama-3.1-Swallow-70B-v0.1](https://huggingface.co/tokyotech-llm/Llama-3.1-Swallow-70B-v0.1) | 368.4 | 379 | 378 | 379 | 351 | 355 |
43
+ | [Meta-Llama-3-70B](https://huggingface.co/meta-llama/Meta-Llama-3-70B) | 334.6 | 353 | 340 | 348 | 314 | 318 |
44
+ | GPT-3.5 | 273.2 | 266 | 250 | 266 | 297 | 287 |
45
+
46
+ ## Limitations
47
+
48
+ The model was developed for research purposes and is not intended for clinical diagnosis.
49
+ It is the users' responsibility to ensure compliance with applicable rules and regulations.
50
+
51
+ ## Contributors
52
+
53
+ Preferred Networks, Inc.
54
+ - Junichiro Iwasawa
55
+ - Wataru Kawakami
56
+ - Keita Suzuki
57
+
58
+ ## Publications
59
+
60
+ We are currently preparing a related blog post and research paper, which will be made available soon to provide further insights and detailed information about the development and capabilities of this model.
61
+
62
+ ## License
63
+
64
+ [Qwen LICENSE](https://huggingface.co/Qwen/Qwen2.5-72B/blob/main/LICENSE)