math-extraction-comp/princeton-nlp__Llama-3-Instruct-8B-SimPO

该数据集是一个包含问题和对应答案以及预测结果的数据集，同时还包括了不同评分系统的评分数据。它分为训练集，用于机器学习模型的训练。数据集包含的字段有：问题(question)、正确答案(gold)、模型预测答案(target)、预测(prediction)、数据子集(subset)、评分系统提取的答案和分数(lighteval-c24870ea_extracted_answer, lighteval-c24870ea_score)、Qwen评分(qwen_score)、另一个评分系统的答案和分数(lighteval-0f21c935_extracted_answer, lighteval-0f21c935_score)、另一个提取的答案和分数(harness_extracted_answer, harness_score)、Qwen提取的答案(qwen_extracted_answer)。

This dataset includes questions, corresponding answers, prediction results, and scores from different scoring systems. It is split into a training set for machine learning model training. The fields contained in the dataset are: question, gold, target, prediction, subset, answers and scores extracted by the lighteval-c24870ea scoring system (lighteval-c24870ea_extracted_answer, lighteval-c24870ea_score), Qwen scores (qwen_score), answers and scores from another scoring system (lighteval-0f21c935_extracted_answer, lighteval-0f21c935_score), answers and scores extracted by another system (harness_extracted_answer, harness_score), and answers extracted by Qwen (qwen_extracted_answer).