Dataset Card for Evaluation run of huggingface/llama-30b
Dataset Summary
Dataset automatically created during the evaluation run of model huggingface/llama-30b on the Open LLM Leaderboard.
The dataset is composed of 64 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 3 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_huggingface__llama-30b.
