Hugging Face's logo

tokyotech-llm
/
Swallow-13b-hf

Text Generation
Transformers
PyTorch
English
Japanese
llama
text-generation-inference
Inference Endpoints
Model card Files Community
2
Taishi-N324 commited on
Commit
f596f12
•
1 Parent(s): 3e8dd06

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +27 -7
README.md CHANGED
@@ -17,6 +17,7 @@ Links to other models can be found in the index.
17
  |Model|Swallow-hf|Swallow-instruct-hf|
18
  |---|---|---|
19
  |7B| [Link](https://huggingface.co/tokyotech-llm/Swallow-7b-hf) | [Link](https://huggingface.co/tokyotech-llm/Swallow-7b-instruct-hf)|
 
20
  |13B| [Link](https://huggingface.co/tokyotech-llm/Swallow-13b-hf) | [Link](https://huggingface.co/tokyotech-llm/Swallow-13b-instruct-hf)|
21
  |70B| [Link](https://huggingface.co/tokyotech-llm/Swallow-70b-hf) | [Link](https://huggingface.co/tokyotech-llm/Swallow-70b-instruct-hf)|
22
 
@@ -27,7 +28,7 @@ Links to other models can be found in the index.
27
  |13B| [Link](https://huggingface.co/tokyotech-llm/Swallow-13b-NVE-hf) | Coming Soon |
28
  |70B| [Link](https://huggingface.co/tokyotech-llm/Swallow-70b-NVE-hf) | [Link](https://huggingface.co/tokyotech-llm/Swallow-70b-NVE-instruct-hf)|
29
 
30
- We released the 7B and 70B models without vocabulary expansion on January 26th, 2024. The 13B model was released on February 4th, 2024, and its instruction-tuned version is coming soon.
31
 
32
 
33
  ![logo](./logo.png)
@@ -51,12 +52,31 @@ Read our [blog post](https://zenn.dev/tokyotech_lm/articles/d6cb3a8fdfc907) or o
51
  |Model|Size|JCommonsenseQA|JEMHopQA|NIILC|JSQuAD|XL-Sum|MGSM|WMT20-en-ja|WMT20-ja-en|
52
  |---|---|---|---|---|---|---|---|---|---|
53
  | | |4-shot|4-shot|4-shot|4-shot|1-shot|4-shot|4-shot|4-shot|
54
- |Llama 2|7B|0.3852|0.4240|0.3410|0.7917|0.1905|0.0760|0.1783|0.1738|
55
- |Swallow|7B|0.4808|0.5078|0.5968|0.8573|0.1830|0.1240|0.2510|0.1511|
56
- |Llama 2|13B|0.6997|0.4415|0.4170|0.8533|0.2139|0.1320|0.2146|0.1982|
57
- |Swallow|13B|0.7837|0.5063|0.6398|0.9005|0.2168|0.2040|0.2720|0.1771|
58
- |Llama 2|70B|0.8686|0.4656|0.5256|0.9080|**0.2361**|0.3560|0.2643|**0.2398**|
59
- |Swallow|70B|**0.9348**|**0.6290**|**0.6960**|**0.9176**|0.2266|**0.4840**|**0.3043**|0.2298|
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
60
 
61
  ## Usage
62
 
 
17
  |Model|Swallow-hf|Swallow-instruct-hf|
18
  |---|---|---|
19
  |7B| [Link](https://huggingface.co/tokyotech-llm/Swallow-7b-hf) | [Link](https://huggingface.co/tokyotech-llm/Swallow-7b-instruct-hf)|
20
+ |7B-Plus| [Link](https://huggingface.co/tokyotech-llm/Swallow-7b-plus-hf) | Coming Soon |
21
  |13B| [Link](https://huggingface.co/tokyotech-llm/Swallow-13b-hf) | [Link](https://huggingface.co/tokyotech-llm/Swallow-13b-instruct-hf)|
22
  |70B| [Link](https://huggingface.co/tokyotech-llm/Swallow-70b-hf) | [Link](https://huggingface.co/tokyotech-llm/Swallow-70b-instruct-hf)|
23
 
 
28
  |13B| [Link](https://huggingface.co/tokyotech-llm/Swallow-13b-NVE-hf) | Coming Soon |
29
  |70B| [Link](https://huggingface.co/tokyotech-llm/Swallow-70b-NVE-hf) | [Link](https://huggingface.co/tokyotech-llm/Swallow-70b-NVE-instruct-hf)|
30
 
31
+ We released the 7B and 70B models without vocabulary expansion on January 26th, 2024. The 13B model was released on February 4th, 2024, and its instruction-tuned version is coming soon. Swallow-7B-Plus is a model that has been trained with a larger number of Japanese tokens compared to Swallow-7B and its release date is March 2nd, 2024.
32
 
33
 
34
  ![logo](./logo.png)
 
52
  |Model|Size|JCommonsenseQA|JEMHopQA|NIILC|JSQuAD|XL-Sum|MGSM|WMT20-en-ja|WMT20-ja-en|
53
  |---|---|---|---|---|---|---|---|---|---|
54
  | | |4-shot|4-shot|4-shot|4-shot|1-shot|4-shot|4-shot|4-shot|
55
+ | Llama 2 | 7B | 0.3852 | 0.4240 | 0.3410 | 0.7917 | 0.1905 | 0.0760 | 0.1783 | 0.1738 |
56
+ | Swallow | 7B | 0.4808 | 0.5078 | 0.5968 | 0.8573 | 0.1830 | 0.1240 | 0.2510 | 0.1511 |
57
+ | Swallow-Plus | 7B | 0.5478 | 0.5493 | 0.6030 | 0.8544 | 0.1806 | 0.1360 | 0.2568 | 0.1441 |
58
+ | Swallow-NVE | 7B | 0.5433 | 0.5425 | 0.5729 | 0.8684 | 0.2117 | 0.1200 | 0.2405 | 0.1512 |
59
+ | Llama 2 | 13B | 0.6997 | 0.4415 | 0.4170 | 0.8533 | 0.2139 | 0.1320 | 0.2146 | 0.1982 |
60
+ | Swallow | 13B | 0.7837 | 0.5063 | 0.6398 | 0.9005 | 0.2168 | 0.2040 | 0.2720 | 0.1771 |
61
+ | Swallow-NVE | 13B | 0.7712 | 0.5438 | 0.6351 | 0.9030 | 0.2294 | 0.2120 | 0.2735 | 0.1817 |
62
+ | Llama 2 | 70B | 0.8686 | 0.4656 | 0.5256 | 0.9080 | 0.2361 | 0.3560 | 0.2643 | **0.2398** |
63
+ | Swallow | 70B | 0.9348 | **0.6290** | 0.6960 | 0.9176 | 0.2266 | **0.4840** | **0.3043** | 0.2298 |
64
+ | Swallow-NVE | 70B | **0.9410** | 0.5759 | **0.7024** | **0.9254** | **0.2758** | 0.4720 | 0.3042 | 0.2322 |
65
+ ### English version
66
+
67
+ |Model|Size|OpenBookQA|TriviaQA|HellaSwag|SQuAD2.0|XWINO|GSM8K|
68
+ |---|---|---|---|---|---|---|---|
69
+ | | |8-shot|8-shot|8-shot|8-shot|8-shot|8-shot|
70
+ | Llama 2 | 7B | 0.3580 | 0.6265 | 0.5860 | 0.3207 | 0.9049 | 0.1410 |
71
+ | Swallow | 7B | 0.3180 | 0.4836 | 0.5308 | 0.3125 | 0.8817 | 0.1130 |
72
+ | Swallow-Plus | 7B | 0.3280 | 0.4558 | 0.5259 | 0.3134 | 0.8929 | 0.1061 |
73
+ | Swallow-NVE | 7B | 0.3180 | 0.5079 | 0.5329 | 0.2919 | 0.8817 | 0.0986 |
74
+ | Llama 2 | 13B | 0.3760 | 0.7255 | 0.6148 | 0.3681 | 0.9140 | 0.2403 |
75
+ | Swallow | 13B | 0.3500 | 0.5852 | 0.5660 | 0.3406 | 0.9075 | 0.2039 |
76
+ | Swallow-NVE | 13B | 0.3460 | 0.6025 | 0.5700 | 0.3478 | 0.9006 | 0.1751 |
77
+ | Llama 2 | 70B | **0.4280** | **0.8239** | **0.6742** | **0.3770** | **0.9290** | **0.5284** |
78
+ | Swallow | 70B | 0.4220 | 0.7756 | 0.6458 | 0.3745 | 0.9204 | 0.4867 |
79
+ | Swallow-NVE | 70B | 0.4240 | 0.7817 | 0.6439 | 0.3451 | 0.9256 | 0.4943 |
80
 
81
  ## Usage
82