mlfoundations-dev/embedding_apply_sharegpt

该数据集包含与指令种子（instruction_seed）、来源（source）相关的文本数据，以及一系列浮点数序列的嵌入表示（embeddings）。此外，还包括正面评分的平均值（mean_positive_score）、负面评分的平均值（mean_negative_score）以及差异分数（difference_score）。数据集分为训练集，共有4000个示例，数据大小约为35093.12MB。

The dataset includes text data related to instruction seeds (instruction_seed), sources (source), and a sequence of floating-point embeddings (embeddings). Additionally, it contains the mean of positive scores (mean_positive_score), the mean of negative scores (mean_negative_score), and the difference score (difference_score). The dataset is split into a training set with a total of 4000 examples, with a size of approximately 35093.12MB.