Dataset Card for Evaluation run of 1000
Dataset Summary
Dataset automatically created during the evaluation run of model 1000 on the Open LLM Leaderboard.
The dataset is composed of 3074 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 903 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 latest… See the full description on the dataset page: https://huggingface.co/datasets/HuggingFaceBR4/thomwolf-third-llama.
