---
library_name: transformers
license: mit
base_model: MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: nli-cross-encoder-roberta
  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. -->

# nli-cross-encoder-roberta

This model is a fine-tuned version of [MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7](https://huggingface.co/MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4859
- Accuracy: 0.9448
- F1 Macro: 0.9469

## 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: 16
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 8
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|
| 0.1806        | 1.0   | 211  | 0.3069          | 0.9088   | 0.9134   |
| 0.1021        | 2.0   | 422  | 0.1795          | 0.9530   | 0.9544   |
| 0.0343        | 3.0   | 633  | 0.4396          | 0.9365   | 0.9389   |
| 0.0182        | 4.0   | 844  | 0.4025          | 0.9475   | 0.9496   |
| 0.0047        | 5.0   | 1055 | 0.4674          | 0.9420   | 0.9441   |
| 0.0014        | 6.0   | 1266 | 0.4457          | 0.9448   | 0.9469   |
| 0.0049        | 7.0   | 1477 | 0.4835          | 0.9448   | 0.9469   |
| 0.0004        | 8.0   | 1688 | 0.4859          | 0.9448   | 0.9469   |


### Framework versions

- Transformers 4.56.1
- Pytorch 2.6.0+cu124
- Datasets 4.0.0
- Tokenizers 0.22.0