Dataset Card for Evaluation run of togethercomputer/Llama-2-7B-32K-Instruct
Dataset Summary
Dataset automatically created during the evaluation run of model togethercomputer/Llama-2-7B-32K-Instruct 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_togethercomputer__Llama-2-7B-32K-Instruct.