search

ManifoldRG / MultiNet

MIT License
10 stars 1 forks source link

Identify tasks for each dataset #47

Open pranavguru opened 3 months ago

pranavguru commented 3 months ago

Identify tasks for each of the datasets to evaluate models on. For eg: Fineweb-edu is a large text corpus. What text-based task should a model be evaluated on after being fine-tuned on Fineweb-edu?

pranavguru commented 3 months ago

Done (https://docs.google.com/document/d/1oJkmtev7s6K-PqV5Ium9XVGC2wPcVxBUhvGj41BWlW8/edit), pending feedback from Benchmark experts and authors