Update README.md
Browse files
README.md
CHANGED
@@ -56,7 +56,7 @@ unfiltered content from the internet, which is far from neutral. As the openAI t
|
|
56 |
> that require the generated text to be true.
|
57 |
>
|
58 |
> Additionally, language models like GPT-2 reflect the biases inherent to the systems they were trained on, so we do
|
59 |
-
> not recommend that they be deployed into systems that interact with humans
|
60 |
> study of biases relevant to the intended use-case. We found no statistically significant difference in gender, race,
|
61 |
> and religious bias probes between 774M and 1.5B, implying all versions of GPT-2 should be approached with similar
|
62 |
> levels of caution around use cases that are sensitive to biases around human attributes.
|
|
|
56 |
> that require the generated text to be true.
|
57 |
>
|
58 |
> Additionally, language models like GPT-2 reflect the biases inherent to the systems they were trained on, so we do
|
59 |
+
> not recommend that they be deployed into systems that interact with humans unless the deployers first carry out a
|
60 |
> study of biases relevant to the intended use-case. We found no statistically significant difference in gender, race,
|
61 |
> and religious bias probes between 774M and 1.5B, implying all versions of GPT-2 should be approached with similar
|
62 |
> levels of caution around use cases that are sensitive to biases around human attributes.
|