Spaces:
Sleeping
Sleeping
Oskar van der Wal
commited on
Fix colors in example
Browse files- description.md +1 -1
description.md
CHANGED
|
@@ -2,4 +2,4 @@
|
|
| 2 |
|
| 3 |
GPT-2 is a language model which can score how likely it is that some text is a valid English sentence: not only grammaticality, but also the 'meaning' of the sentence is part of this score. CrowS-Pairs is a dataset with pairs of more and less stereotypical examples for different social groups (e.g., gender and nationality stereotypes). We sample 10 random pairs from CrowS-Pairs and show whether the stereotypical example gets a higher score ('is more likely'). If GPT-2 systematically prefers the stereotypical examples, it has probably learnt these stereotypes from the training data.
|
| 4 |
|
| 5 |
-
The colors indicate whether the <font color=#
|
|
|
|
| 2 |
|
| 3 |
GPT-2 is a language model which can score how likely it is that some text is a valid English sentence: not only grammaticality, but also the 'meaning' of the sentence is part of this score. CrowS-Pairs is a dataset with pairs of more and less stereotypical examples for different social groups (e.g., gender and nationality stereotypes). We sample 10 random pairs from CrowS-Pairs and show whether the stereotypical example gets a higher score ('is more likely'). If GPT-2 systematically prefers the stereotypical examples, it has probably learnt these stereotypes from the training data.
|
| 4 |
|
| 5 |
+
The colors indicate whether the <font color=#00ffff>stereotypical</font> or the <font color=#ff00ff>less stereotypical</font> examples gets the higher score, the intensity of the color how strong the preference is.
|