Mitsuki-Sakamoto/hh-rlhf-reward-model-deberta-v3-large-v2-harmless-2-original_mix_90_random_seed_3

--- dataset_info: features: - name: chosen dtype: string - name: rejected dtype: string - name: reward_model_prompt_format dtype: string - name: reward_1 dtype: float64 - name: reward_2 dtype: float64 - name: index dtype: int64 - name: generate_format_dict dtype: string - name: gen_kwargs struct: - name: do_sample dtype: bool - name: max_new_tokens dtype: int64 - name: pad_token_id dtype: int64 - name: top_k dtype: int64 - name: top_p dtype: float64 - name: n_samples dtype: int64 - name: reject_select dtype: string splits: - name: train num_bytes: 99998874.57667443 num_examples: 42536 - name: test num_bytes: 5542190.321583045 num_examples: 2311 download_size: 65009169 dataset_size: 105541064.89825748 configs: - config_name: default data_files: - split: train path: data/train-* - split: test path: data/test-* ---

原始信息汇总

数据集概述

数据集信息

特征

• chosen: 类型为字符串
• rejected: 类型为字符串
• reward_model_prompt_format: 类型为字符串
• reward_1: 类型为浮点数 (float64)
• reward_2: 类型为浮点数 (float64)
• index: 类型为整数 (int64)
• generate_format_dict: 类型为字符串
• gen_kwargs: 结构化数据
  • do_sample: 类型为布尔值 (bool)
  • max_new_tokens: 类型为整数 (int64)
  • pad_token_id: 类型为整数 (int64)
  • top_k: 类型为整数 (int64)
  • top_p: 类型为浮点数 (float64)
• n_samples: 类型为整数 (int64)
• reject_select: 类型为字符串

数据分割

• train: 包含42536个样本，占用99998874.57667443字节
• test: 包含2311个样本，占用5542190.321583045字节

数据集大小

• 下载大小: 65009169字节
• 总大小: 105541064.89825748字节

配置

• config_name: default
  • data_files:
    • train: 路径为 data/train-*
    • test: 路径为 data/test-*