OALL/details_princeton-nlp__Llama-3-Base-8B-SFT-IPO

Dataset Card for Evaluation run of princeton-nlp/Llama-3-Base-8B-SFT-IPO Dataset automatically created during the evaluation run of model princeton-nlp/Llama-3-Base-8B-SFT-IPO. The dataset is composed of 136 configuration, each one coresponding to one of the evaluated task. The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the… See the full description on the dataset page: https://huggingface.co/datasets/OALL/details_princeton-nlp__Llama-3-Base-8B-SFT-IPO.