Hi guys!
I’m writing an academic paper on Filter Functions in LLMs.
For evaluation purposes I need to check for the ability to filter out certain code libraries, and I think the best way to do this would be to get a dataset with code requests (“hey can you write a program that does X?”), specifically requests for neural nets with pytorch\/tensorflow. <\/p>\n
Just to make clear – I do not need to train any model on these, just to run them through the LLM with\/out the filter.<\/p>\n
Example – “Hey can you build a neural network that classifies semantics of tweets?”
I don’t need anything too complicated<\/p>\n
I’ve searched standard datasets on huggingface\/google but haven’t found any with enough samples.
Any ideas?
Any help would be much appreciated and I’d love to answer any questions about the research itself. <\/p>\n
Thanks!<\/p>\n<\/div>\n
submitted by \/u\/AltivoTheHorseX <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
