grok-1
grok-1 copied to clipboard
Dataset
What dataset was used to train grok? Will it be open-sourced? How are you going to make model/dataset up to date? Can you provide an example of dataset?
This issue is related to https://github.com/xai-org/grok-1/issues/43 and https://github.com/xai-org/grok-1/issues/41
I'm also interested in a description of the training data.
yes. if the training data isn't released (at least the titles of the data items) then it's not technically completely open source.