Update README benchmark scores only
Browse files
README.md
CHANGED
|
@@ -123,26 +123,3 @@ This code repository is licensed under the [MIT License](LICENSE). The use of My
|
|
| 123 |
## 6. Contact
|
| 124 |
If you have any questions, please raise an issue on our GitHub repository or contact us at contact@MyAwesomeModel.ai.
|
| 125 |
```
|
| 126 |
-
|
| 127 |
-
|
| 128 |
-
## Detailed Evaluation Results
|
| 129 |
-
|
| 130 |
-
Selected checkpoint: `step_1000` with `eval_accuracy` = **0.710**.
|
| 131 |
-
|
| 132 |
-
| Benchmark | Score |
|
| 133 |
-
|---|---:|
|
| 134 |
-
| math_reasoning | 0.550 |
|
| 135 |
-
| logical_reasoning | 0.819 |
|
| 136 |
-
| common_sense | 0.736 |
|
| 137 |
-
| reading_comprehension | 0.700 |
|
| 138 |
-
| question_answering | 0.607 |
|
| 139 |
-
| text_classification | 0.828 |
|
| 140 |
-
| sentiment_analysis | 0.792 |
|
| 141 |
-
| code_generation | 0.650 |
|
| 142 |
-
| creative_writing | 0.610 |
|
| 143 |
-
| dialogue_generation | 0.644 |
|
| 144 |
-
| summarization | 0.767 |
|
| 145 |
-
| translation | 0.804 |
|
| 146 |
-
| knowledge_retrieval | 0.676 |
|
| 147 |
-
| instruction_following | 0.758 |
|
| 148 |
-
| safety_evaluation | 0.739 |
|
|
|
|
| 123 |
## 6. Contact
|
| 124 |
If you have any questions, please raise an issue on our GitHub repository or contact us at contact@MyAwesomeModel.ai.
|
| 125 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|