---
license: mit
base_model: mikhail-panzo/malay_full_checkpoint
tags:
- generated_from_trainer
model-index:
- name: fil_checkpoint
  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. -->

# fil_checkpoint

This model is a fine-tuned version of [mikhail-panzo/malay_full_checkpoint](https://huggingface.co/mikhail-panzo/malay_full_checkpoint) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4735

## 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: 1e-06
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2000
- training_steps: 3000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 0.7579        | 6.76  | 500  | 0.7164          |
| 0.6196        | 13.51 | 1000 | 0.5662          |
| 0.5622        | 20.27 | 1500 | 0.5077          |
| 0.5341        | 27.03 | 2000 | 0.4858          |
| 0.52          | 33.78 | 2500 | 0.4772          |
| 0.5233        | 40.54 | 3000 | 0.4735          |


### Framework versions

- Transformers 4.40.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2