xPXXX/tevatron_wikipedia-nq_sample100

Name: xPXXX/tevatron_wikipedia-nq_sample100
Creator: xPXXX
Published: 2024-04-17 22:32:21
License: 暂无描述

Hugging Face2024-04-17 更新2024-06-12 收录

下载链接：

https://hf-mirror.com/datasets/xPXXX/tevatron_wikipedia-nq_sample100

下载链接

链接失效反馈

官方服务：

资源简介：

--- dataset_info: - config_name: default features: - name: query_id dtype: string - name: query dtype: string - name: answers list: string - name: positive_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: negative_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string splits: - name: train num_bytes: 6516114.769199276 num_examples: 100 download_size: 3750358 dataset_size: 6516114.769199276 - config_name: finetune_llama2_no_rag features: - name: query_id dtype: string - name: query dtype: string - name: answers list: string - name: positive_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: negative_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: finetune_llama2_no_rag_response dtype: string splits: - name: train num_bytes: 6539762 num_examples: 100 download_size: 3774951 dataset_size: 6539762 - config_name: finetune_llama2_rag features: - name: query_id dtype: string - name: query dtype: string - name: answers list: string - name: positive_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: negative_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: finetune_llama2_rag_response dtype: string - name: retrieval sequence: string splits: - name: train num_bytes: 6781481 num_examples: 100 download_size: 3872487 dataset_size: 6781481 - config_name: gpt3_no_rag features: - name: query_id dtype: string - name: query dtype: string - name: answers list: string - name: positive_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: negative_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: gpt3_no_rag_response dtype: string splits: - name: train num_bytes: 6530548 num_examples: 100 download_size: 3769177 dataset_size: 6530548 - config_name: gpt3_rag features: - name: query_id dtype: string - name: query dtype: string - name: answers list: string - name: positive_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: negative_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: gpt3_rag_response dtype: string - name: retrieval sequence: string splits: - name: train num_bytes: 6794643 num_examples: 100 download_size: 3882537 dataset_size: 6794643 - config_name: gpt4_no_rag features: - name: query_id dtype: string - name: query dtype: string - name: answers list: string - name: positive_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: negative_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: gpt4_no_rag_response dtype: string splits: - name: train num_bytes: 6565730 num_examples: 100 download_size: 3789108 dataset_size: 6565730 - config_name: gpt4_rag features: - name: query_id dtype: string - name: query dtype: string - name: answers list: string - name: positive_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: negative_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: gpt4_rag_response dtype: string - name: retrieval sequence: string splits: - name: train num_bytes: 6801342 num_examples: 100 download_size: 3885780 dataset_size: 6801342 - config_name: llama2_no_rag features: - name: query_id dtype: string - name: query dtype: string - name: answers list: string - name: positive_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: negative_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: llama2_no_rag_response dtype: string splits: - name: train num_bytes: 6550530 num_examples: 100 download_size: 3781797 dataset_size: 6550530 - config_name: llama2_rag features: - name: query_id dtype: string - name: query dtype: string - name: answers list: string - name: positive_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: negative_passages list: - name: docid dtype: string - name: text dtype: string - name: title dtype: string - name: llama2_rag_response dtype: string - name: retrieval sequence: string splits: - name: train num_bytes: 6789056 num_examples: 100 download_size: 3877897 dataset_size: 6789056 configs: - config_name: default data_files: - split: train path: data/train-* - config_name: finetune_llama2_no_rag data_files: - split: train path: finetune_llama2_no_rag/train-* - config_name: finetune_llama2_rag data_files: - split: train path: finetune_llama2_rag/train-* - config_name: gpt3_no_rag data_files: - split: train path: gpt3_no_rag/train-* - config_name: gpt3_rag data_files: - split: train path: gpt3_rag/train-* - config_name: gpt4_no_rag data_files: - split: train path: gpt4_no_rag/train-* - config_name: gpt4_rag data_files: - split: train path: gpt4_rag/train-* - config_name: llama2_no_rag data_files: - split: train path: llama2_no_rag/train-* - config_name: llama2_rag data_files: - split: train path: llama2_rag/train-* ---

提供机构：

xPXXX

原始信息汇总

数据集概述

数据集配置

默认配置 (`default`)

特征:
- query_id: 字符串
- query: 字符串
- answers: 字符串列表
- positive_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- negative_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
分割:
- train:
  - 字节数: 6516114.769199276
  - 样本数: 100
下载大小: 3750358
数据集大小: 6516114.769199276

微调 Llama2 无 RAG (`finetune_llama2_no_rag`)

特征:
- query_id: 字符串
- query: 字符串
- answers: 字符串列表
- positive_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- negative_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- finetune_llama2_no_rag_response: 字符串
分割:
- train:
  - 字节数: 6539762
  - 样本数: 100
下载大小: 3774951
数据集大小: 6539762

微调 Llama2 RAG (`finetune_llama2_rag`)

特征:
- query_id: 字符串
- query: 字符串
- answers: 字符串列表
- positive_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- negative_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- finetune_llama2_rag_response: 字符串
- retrieval: 字符串序列
分割:
- train:
  - 字节数: 6781481
  - 样本数: 100
下载大小: 3872487
数据集大小: 6781481

GPT-3 无 RAG (`gpt3_no_rag`)

特征:
- query_id: 字符串
- query: 字符串
- answers: 字符串列表
- positive_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- negative_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- gpt3_no_rag_response: 字符串
分割:
- train:
  - 字节数: 6530548
  - 样本数: 100
下载大小: 3769177
数据集大小: 6530548

GPT-3 RAG (`gpt3_rag`)

特征:
- query_id: 字符串
- query: 字符串
- answers: 字符串列表
- positive_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- negative_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- gpt3_rag_response: 字符串
- retrieval: 字符串序列
分割:
- train:
  - 字节数: 6794643
  - 样本数: 100
下载大小: 3882537
数据集大小: 6794643

GPT-4 无 RAG (`gpt4_no_rag`)

特征:
- query_id: 字符串
- query: 字符串
- answers: 字符串列表
- positive_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- negative_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- gpt4_no_rag_response: 字符串
分割:
- train:
  - 字节数: 6565730
  - 样本数: 100
下载大小: 3789108
数据集大小: 6565730

GPT-4 RAG (`gpt4_rag`)

特征:
- query_id: 字符串
- query: 字符串
- answers: 字符串列表
- positive_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- negative_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- gpt4_rag_response: 字符串
- retrieval: 字符串序列
分割:
- train:
  - 字节数: 6801342
  - 样本数: 100
下载大小: 3885780
数据集大小: 6801342

Llama2 无 RAG (`llama2_no_rag`)

特征:
- query_id: 字符串
- query: 字符串
- answers: 字符串列表
- positive_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- negative_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- llama2_no_rag_response: 字符串
分割:
- train:
  - 字节数: 6550530
  - 样本数: 100
下载大小: 3781797
数据集大小: 6550530

Llama2 RAG (`llama2_rag`)

特征:
- query_id: 字符串
- query: 字符串
- answers: 字符串列表
- positive_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- negative_passages: 列表
  - docid: 字符串
  - text: 字符串
  - title: 字符串
- llama2_rag_response: 字符串
- retrieval: 字符串序列
分割:
- train:
  - 字节数: 6789056
  - 样本数: 100
下载大小: 3877897
数据集大小: 6789056

5,000+

优质数据集

54 个

任务类型

进入经典数据集

xPXXX/tevatron_wikipedia-nq_sample100

数据集概述

数据集配置

默认配置 (default)

微调 Llama2 无 RAG (finetune_llama2_no_rag)

微调 Llama2 RAG (finetune_llama2_rag)

GPT-3 无 RAG (gpt3_no_rag)

GPT-3 RAG (gpt3_rag)

GPT-4 无 RAG (gpt4_no_rag)

GPT-4 RAG (gpt4_rag)

Llama2 无 RAG (llama2_no_rag)

Llama2 RAG (llama2_rag)