---
tags:
- generated_from_trainer
datasets:
- wikipedia
base_model: robingeibel/reformer-finetuned-big_patent-wikipedia-arxiv-16384
model-index:
- name: reformer-finetuned-big_patent-wikipedia-arxiv-16384
  results: []
---

<!-- 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. -->

# reformer-finetuned-big_patent-wikipedia-arxiv-16384

This model is a fine-tuned version of [robingeibel/reformer-finetuned-big_patent-wikipedia-arxiv-16384](https://huggingface.co./robingeibel/reformer-finetuned-big_patent-wikipedia-arxiv-16384) on the wikipedia dataset.
It achieves the following results on the evaluation set:
- Loss: 6.5256

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 2.5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 8.0368        | 1.0   | 3785  | 6.7392          |
| 6.7992        | 2.0   | 7570  | 6.5576          |
| 6.6926        | 3.0   | 11355 | 6.5256          |


### Framework versions

- Transformers 4.20.1
- Pytorch 1.12.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1