ferrazzipietro/unLS_Llama-2-7b-hf_ncbi_disease_NoQuant_64_64_0.05_32_BestF1

该数据集主要用于命名实体识别任务，包含id、tokens、ner_tags、input_ids、attention_mask、labels、sentence、predictions和ground_truth_labels等字段。ner_tags字段用于标注疾病相关的实体，标签类别包括O（非实体）、B-Disease（疾病实体的开始）和I-Disease（疾病实体的内部）。数据集仅包含一个测试集，包含924个样本，文件大小为1385467字节。

This dataset is primarily used for Named Entity Recognition tasks, containing multiple features such as id, tokens, ner_tags, etc. The ner_tags feature is used to identify disease entities, including three labels: O (non-entity), B-Disease (beginning of disease entity), and I-Disease (inside of disease entity). The dataset is divided into multiple parts, with the test part containing 924 samples.