yuruny/sokoban_state_Markovian_S6-6_B1-1_D6-6_SFT_dataset

--- dataset_info: features: - name: prompt list: - name: content dtype: string - name: role dtype: string - name: answer dtype: string splits: - name: train num_bytes: 13469328 num_examples: 15240 download_size: 121295 dataset_size: 13469328 configs: - config_name: default data_files: - split: train path: data/train-* ---This dataset contains two core features: prompt and answer. The prompt is a list that includes two fields, content and role, both of which are of string type. The answer is also a string-type value. The dataset includes a split named "train", which contains 15,240 samples, with a total size of 13,469,328 bytes and a download size of 121,295 bytes. The configuration information specifies the data file path for the training split.