README.md

---
language:
- en
multilinguality:
- monolingual
size_categories:
- 100K<n<1M
task_categories:
- feature-extraction
- sentence-similarity
pretty_name: ELI5
tags:
- sentence-transformers
dataset_info:
  config_name: pair
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  splits:
  - name: train
    num_bytes: 173287519
    num_examples: 325475
  download_size: 112893890
  dataset_size: 173287519
configs:
- config_name: pair
  data_files:
  - split: train
    path: pair/train-*
---

# Dataset Card for ELI5

This dataset is a collection of question-answer pairs, collected from the Explain Like I'm 5 subreddit. See [ELI5](https://huggingface.co/datasets/eli5) for additional information.
This dataset can be used directly with Sentence Transformers to train embedding models.

## Dataset Subsets

### `pair` subset

* Columns: "question", "answer"
* Column types: `str`, `str`
* Examples:
    ```python
    {
      'question': 'Why chemical weapons considered more indiscriminate than conventional weapons?',
      'answer': "Well, any large-scale ordinance is indiscriminate. The problem particularly with Chemical weapons is that, especially with those that are gas based, is that the actual range of the weapon is much larger than the blast radius. The Chemical residue can remain in the area for a long period of time, it can taint and damage water and food supplies, it can be carried on clothing, such that you could drop it on a house full of terrorists, but if there is an orphanage upwind, they are going to get some of it as well. With conventional ordinance, you can target and turn that building full of terrorists into rubble, and thanks to years of testing, we know more or less a good idea of the collateral damage. With a chemical warhead, its not one building - its anything and everything in that area. There is nothing pinpoint about a chemical weapon system - it can and will spread beyond the impact zone That is what makes it (imho) more 'indescriminate' than conventional weaponry.",
    }
    ```
* Collection strategy: Reading the ELI5 dataset from [embedding-training-data](https://huggingface.co/datasets/sentence-transformers/embedding-training-data).
* Deduplified: No