---
license: mit
tags:
- generated_from_trainer
model-index:
- name: roberta-base-mnli_AppE
  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. -->

# roberta-base-mnli_AppE

This model is a fine-tuned version of [WillHeld/roberta-base-mnli](https://huggingface.co/WillHeld/roberta-base-mnli) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7433
- Acc: 0.8622

## 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: 2e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Acc    |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 0.3562        | 0.17  | 2000  | 0.4166          | 0.8524 |
| 0.3414        | 0.33  | 4000  | 0.4399          | 0.8476 |
| 0.3308        | 0.5   | 6000  | 0.3951          | 0.8561 |
| 0.3306        | 0.67  | 8000  | 0.4152          | 0.8488 |
| 0.3295        | 0.83  | 10000 | 0.3917          | 0.8575 |
| 0.3288        | 1.0   | 12000 | 0.4095          | 0.8581 |
| 0.229         | 1.17  | 14000 | 0.4420          | 0.8611 |
| 0.2307        | 1.33  | 16000 | 0.4876          | 0.8561 |
| 0.2334        | 1.5   | 18000 | 0.4496          | 0.8577 |
| 0.2392        | 1.67  | 20000 | 0.4188          | 0.8608 |
| 0.233         | 1.83  | 22000 | 0.4493          | 0.8578 |
| 0.2343        | 2.0   | 24000 | 0.4278          | 0.8603 |
| 0.163         | 2.17  | 26000 | 0.5700          | 0.8635 |
| 0.1657        | 2.33  | 28000 | 0.5323          | 0.8561 |
| 0.1652        | 2.5   | 30000 | 0.5047          | 0.8589 |
| 0.1654        | 2.67  | 32000 | 0.5082          | 0.8590 |
| 0.1659        | 2.83  | 34000 | 0.5076          | 0.8631 |
| 0.1654        | 3.0   | 36000 | 0.5036          | 0.8612 |
| 0.1195        | 3.17  | 38000 | 0.6221          | 0.8600 |
| 0.1194        | 3.33  | 40000 | 0.6467          | 0.8575 |
| 0.1205        | 3.5   | 42000 | 0.6742          | 0.8578 |
| 0.1209        | 3.67  | 44000 | 0.6537          | 0.8580 |
| 0.1234        | 3.83  | 46000 | 0.6385          | 0.8580 |
| 0.122         | 4.0   | 48000 | 0.6104          | 0.8628 |
| 0.0937        | 4.17  | 50000 | 0.7261          | 0.8603 |
| 0.0941        | 4.33  | 52000 | 0.7634          | 0.8607 |
| 0.0926        | 4.5   | 54000 | 0.7737          | 0.8617 |
| 0.093         | 4.67  | 56000 | 0.7577          | 0.8633 |
| 0.0918        | 4.83  | 58000 | 0.7535          | 0.8631 |
| 0.0935        | 5.0   | 60000 | 0.7433          | 0.8622 |


### Framework versions

- Transformers 4.24.0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.12.1