Dataset used for training?
#7
by
lanortha
- opened
What was the size of the dataset used to train this model? Was it entirely based off the referenced paper?
We use a filtered version of the referenced dataset; the dataset is being improved and enlarged, further details of the updated version will be released along with v2 of our model within the next few weeks/months.
Thanks.
One other question ... the results ... what are they exactly? What do the values correspond to? How do I interpret them? (Apologies, I'm fairly new to using these types of models).
The output is a list of three probabilities indicating how likely the input text expresses left-wing (index 0), centrist (index 1), or right-wing (index 2) political views.