---
library_name: transformers
license: apache-2.0
base_model: distilbert/distilbert-base-uncased
tags:
- generated_from_trainer
model-index:
- name: question_answering_model
  results: []
datasets:
- rajpurkar/squad
pipeline_tag: question-answering
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# question_answering_model

This model is a fine-tuned version of [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased) on SQuAD.
It achieves the following results on the evaluation set:
- Loss: 1.1407

## Model description

Question and Answering model fine-tuned on SQuAD.

## Intended uses & limitations

Educational demo of extractive QA with transformers. Not for production, medical, legal, or safety-critical use. 

## Citation Information

```@inproceedings{rajpurkar-etal-2016-squad,
    title = "{SQ}u{AD}: 100,000+ Questions for Machine Comprehension of Text",
    author = "Rajpurkar, Pranav  and
      Zhang, Jian  and
      Lopyrev, Konstantin  and
      Liang, Percy",
    editor = "Su, Jian  and
      Duh, Kevin  and
      Carreras, Xavier",
    booktitle = "Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2016",
    address = "Austin, Texas",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/D16-1264",
    doi = "10.18653/v1/D16-1264",
    pages = "2383--2392",
    eprint={1606.05250},
    archivePrefix={arXiv},
    primaryClass={cs.CL},
}



## Training and evaluation data

Trained on [squad](https://huggingface.co/datasets/rajpurkar/squad) (train).
Evaluated on its validation split.

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 3

### Training results

| Training Loss | Epoch | Step  | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 1.1995        | 1.0   | 5475  | 1.1494          |
| 0.9689        | 2.0   | 10950 | 1.0921          |
| 0.7334        | 3.0   | 16425 | 1.1407          |


### Framework versions

- Transformers 5.0.0.dev0
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1