evaluation
evaluation copied to clipboard
Published
20 hours ago •
bigscience-workshop
bigscience-workshop
Code and Data for Evaluation WG
Results
50
evaluation issues
Sort by
recently updated
recently updated
newest added
related: #79 @jaketae: here you go :) (_I'm having more free time than usual until the next Monday, so please feel free to assign me other tasks._)
use to test generalization to unseen labels; maybe use FLEX?
few_shot
- Evaluated on: GPT Neo There is an error would happen if you want to evaluate gpt-neo related models (e.g. "EleutherAI/gpt-neo-125M") using the original version of transformers. This is due...
