xtestlab-eternalyc-fyi-1
收藏资源简介:
Xtestlab-Eternalyc-Fyi-1是一个合成数据集,由NeMo Data Designer框架生成。该数据集围绕虚构的xtestlab公司及其产品“eternalyc-fyi-1”构建,该产品旨在提供“抑制人类生物衰老”的解决方案。数据集包含1000行数据,共3个字段:evidence_quote(证据引用)、answer(答案)和question(问题),所有字段均为字符串类型。统计显示,evidence_quote字段的唯一值占比为1.5%,answer字段为2.3%,question字段为1.1%,且所有字段均无空值。数据生成过程中采用了7种列配置,包括表达式生成、LLM结构化生成和种子数据集生成等方法。该数据集适用于自然语言处理任务,如问答系统构建、文本生成或合成数据研究,尤其适合需要模拟特定领域(如抗衰老技术)对话或知识库的场景。
Xtestlab-Eternalyc-Fyi-1 is a synthetic dataset generated by the NeMo Data Designer framework. It is built around the fictional xtestlab company and its product eternalyc-fyi-1, which aims to provide a solution for suppressing human biological aging. The dataset contains 1000 rows of data with 3 fields: evidence_quote (evidence citation), answer (answer), and question (question), all of which are string types. Statistics show that the unique value ratios are 1.5% for evidence_quote, 2.3% for answer, and 1.1% for question, with no null values in any field. The data generation process involved 7 column configurations, including methods such as expression generation, LLM-structured generation, and seed dataset generation. This dataset is suitable for natural language processing tasks, such as question-answering system construction, text generation, or synthetic data research, particularly for scenarios requiring simulation of domain-specific dialogues or knowledge bases, such as anti-aging technology.
数据集概述:Xtestlab-Eternalyc-Fyi-1
- 数据集规模:1,000 条记录
- 数据量级:1K < n < 10K
- 数据格式:Parquet 文件
- 配置名称:
data(默认配置) - 标签:synthetic、datadesigner
数据集描述
该数据集为虚构公司 xtestlab 的产品 eternalyc-fyi-1 生成的模拟数据,该产品提供“抑制人类生物衰老”的解决方案。数据集包含 1,000 行纯数据。
数据模式与统计
| 列名 | 类型 | 列类型 | 唯一值比例 | 空值比例 |
|---|---|---|---|---|
evidence_quote |
string |
expression | 1.5% | 0.0% |
answer |
string |
expression | 2.3% | 0.0% |
question |
string |
expression | 1.1% | 0.0% |
生成详情
数据集通过多种列配置生成,共使用 7 种列配置,包括:
- expression:4 列
- llm-structured:1 列
- seed-dataset:2 列
完整配置和元数据分别存储于 builder_config.json 和 metadata.json 文件中。
引用信息
若使用该数据集,建议引用 NeMo Data Designer 项目:
bibtex @misc{nemo-data-designer, author = {The NeMo Data Designer Team, NVIDIA}, title = {NeMo Data Designer: A framework for generating synthetic data from scratch or based on your own seed data}, howpublished = {url{https://github.com/NVIDIA-NeMo/DataDesigner}}, year = 2026, note = {GitHub Repository}, }




