Processing of the data

#1
by IdoAmit198 - opened

Hi,
I want to use this dataset to pre-train GPT-2.
However, according to this official site the dataset should be pre-processed (such as filter non-english, remove duplications, etc).

My question is whether This dataset is already processed and can be used as is, or I should do the required steps by my self?

In addition, if anyone could share an example (code, notebook) to using this dataset in training GPT-2 (maybe with the run_clm.py script?), I would appreciate that!

Thanks in advance everyone :)

Hello,

I'm not sure if you;re still working on it or not, but according to this OpenWebText blogpost it has already some preprocessing included.

Sign up or log in to comment