Update README.md

README.md

@@ -3,6 +3,21 @@ library_name: peft
 datasets:
 - HachiML/databricks-dolly-15k-ja-alpaca-format
 ---
+## JGLUE Score
+I evaluated this model using the following JGLUE tasks. Here are the scores:
+| Task                | stablelm-base-alpha-7b | qlora-ja-2ep(*) | This Model(8ep) | stablelm-instruct-alpha-7b |
+|---------------------|:-----------------:|:----------:|:----------:|:-----------------:|
+| JCOMMONSENSEQA(acc) | 33.42             | 79.17      | 73.36      | 82.22             |
+| JNLI(acc)           | 43.34             | 47.82      | 39.44      | 52.05             |
+| MARC_JA(acc)        | 96.73             | 88.14      | 80.38      | 82.88             |
+| JSQUAD(exact_match) | 70.62             | 29.85      | 19.33      | 63.26             |
+| **Average**         | **61.03**         | **61.25**  | **53.13**  | **70.10**         |
+- Note: Use v0.3 prompt template
+- The JGLUE scores were measured using the following script:
+[Stability-AI/lm-evaluation-harness](https://github.com/Stability-AI/lm-evaluation-harness/tree/jp-stable)
+- The JGLUE scores of Model "stablelm-base-alpha-7b" and "stablelm-instruct-alpha-7b" were referenced from Github above.
+- (*) [HachiML/japanese-stablelm-alpha-7b-instruct-ja-qlora-2ep-v2](https://huggingface.co/HachiML/japanese-stablelm-alpha-7b-instruct-ja-qlora-2ep-v2)
+
 ## Training procedure