iproskurina/bias-in-bios-llama-instruct-seed30

--- dataset_info: features: - name: seed_len dtype: int64 - name: temperature dtype: float64 - name: seed dtype: string - name: human_full dtype: string - name: human_continuation dtype: string - name: model_full dtype: string splits: - name: train num_bytes: 13763439 num_examples: 11334 download_size: 8744054 dataset_size: 13763439 configs: - config_name: default data_files: - split: train path: data/train-* ---This dataset contains text continuations generated by humans and models. The relevant parameters include seed length (seed_len), temperature, seed (seed), full human-written text (human_full), human-written continuation text (human_continuation), and full model-generated text (model_full). The dataset has a split named "train", which includes a total of 11,334 samples with an overall size of 13,763,439 bytes.