Updates to Model Card noting issues with generated answers
Browse filesIt is very easy to generate *extremely* unethical answers (eg plans for genocide) using this model. This should be noted.
README.md
CHANGED
|
@@ -59,6 +59,8 @@ which constitutes a significant part of the StackExchange data,
|
|
| 59 |
most users who answered the survey identified themselves as [White or European, men, between 25 and 34 years old, and based in the US (with a significant part of responders from India).](https://survey.stackoverflow.co/2022/#developer-profile-demographics)
|
| 60 |
- May generate answers that are incorrect or misleading.
|
| 61 |
- May copy answers from the training data verbatim.
|
|
|
|
|
|
|
| 62 |
|
| 63 |
|
| 64 |
### Recommendations
|
|
|
|
| 59 |
most users who answered the survey identified themselves as [White or European, men, between 25 and 34 years old, and based in the US (with a significant part of responders from India).](https://survey.stackoverflow.co/2022/#developer-profile-demographics)
|
| 60 |
- May generate answers that are incorrect or misleading.
|
| 61 |
- May copy answers from the training data verbatim.
|
| 62 |
+
- Contains extremely NSFW data.
|
| 63 |
+
- Suggested answers maybe illegal, unethical and/or distateful.
|
| 64 |
|
| 65 |
|
| 66 |
### Recommendations
|