Hritikbansal

icon indicating a website link https://sites.google.com/view/hbansal

icon company UCLA icon location Los Angeles icon location Generative AI | Multimodal ML | LLM

Results 2 issues of Hritikbansal

OPT Results not matching for HellaSWAG dataset

5 comment

Hi, I tried reproducing the OPT results for various datasets using the [LM-eval-harness](https://github.com/EleutherAI/lm-evaluation-harness) framework. I observe that the OPT **Accuracy** scores **do not** match the ones reported in the Figure...

bug

Regarding the validity of the Human Preference Classifier

1 comment

Hi @tgxs002 , thanks for your work, and making the dataset and classifier open-sourced! As a sanity check, I evaluated your trained HPC on the examples in the training data...