Hacker News new | ask | show | jobs
by janalsncm 6 days ago
If you are training an LLM, you need to tokenize the text before it’s trained on. A lot of time this can be done in parallel with the GPU though.

I have spent way too much time waiting 10-15 minutes tokenizing my training dataset only for the run to crash over some minor bug after that. (If I was smarter, I’d test on a smaller batch first.)
1 comments
antonvs 6 days ago
> … only for the run to crash over some minor bug after that

Ah, Python.

link
janalsncm 6 days ago
An example would be setting batch size too high causing OOM which isn’t really a python problem.
link