110 lines
4.9 KiB
Markdown
110 lines
4.9 KiB
Markdown
---
|
|
language:
|
|
- nl
|
|
license: mit
|
|
tags:
|
|
- trl
|
|
- fietje
|
|
- alignment-handbook
|
|
- sft
|
|
base_model: BramVanroy/fietje-2
|
|
datasets:
|
|
- BramVanroy/ultrachat_200k_dutch
|
|
- BramVanroy/no_robots_dutch
|
|
- BramVanroy/belebele_dutch
|
|
pipeline_tag: text-generation
|
|
inference: false
|
|
model-index:
|
|
- name: fietje-2-instruct
|
|
results: []
|
|
---
|
|
|
|
<p align="center" style="margin:0;padding:0">
|
|
<img src="https://huggingface.co/BramVanroy/fietje-2-instruct/resolve/main/img/fietje-2b-banner-rounded.png" alt="Fietje banner" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
|
|
</p>
|
|
|
|
<div style="margin:auto; text-align:center">
|
|
<h1 style="margin-bottom: 0">Fietje 2 Instruct</h1>
|
|
<em>An open and efficient LLM for Dutch</em>
|
|
</div>
|
|
|
|
<blockquote class="tip" style="padding: 1.5em; border: 0">
|
|
<p align="center" style="text-align: center; margin: 0">
|
|
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2">👱♀️ Base version</a> -
|
|
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2-instruct">🤖 Instruct version</a> (this one) -
|
|
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2-chat">💬 Chat version</a> -
|
|
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2-chat-GGUF">🚀 GGUF of Instruct</a>
|
|
</p>
|
|
</blockquote>
|
|
|
|
This is the instruct version of Fietje, an SFT-tuned (instruction-tuned) variant of [the base model](https://huggingface.co/BramVanroy/fietje-2). Fietje is an adapated version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2), tailored to Dutch text generation by training on 28B tokens. It is small and efficient with a size of 2.7 billion parameters while performing almost on par with more powerful Dutch LLMs of twice its size like [GEITje 7B Ultra](https://huggingface.co/BramVanroy/GEITje-7B-ultra).
|
|
|
|
A thorough description of the creation and evaluation of Fietje as well as usage examples are available in [this Github repository](https://github.com/BramVanroy/fietje).
|
|
|
|
## Citation
|
|
|
|
If you use Fietje or the [CulturaX + Wikipedia filtered subset](https://huggingface.co/datasets/BramVanroy/wikipedia_culturax_dutch) in your work, please cite to the following paper:
|
|
|
|
```bibtex
|
|
@article{vanroy2024fietje,
|
|
author="Vanroy, Bram",
|
|
title="Fietje: An open, efficient LLM for Dutch",
|
|
url="https://www.clinjournal.org/clinj/article/view/213",
|
|
journal="Computational Linguistics in the Netherlands Journal",
|
|
volume="14",
|
|
year="2025",
|
|
pages="473--504"
|
|
}
|
|
```
|
|
|
|
## Intended uses & limitations
|
|
|
|
The same limitations as [phi-2](https://huggingface.co/microsoft/phi-2#limitations-of-phi-2), and LLMs in general, apply here. LLMs hallucinate, make mistakes, and should not be trusted. Use at your own risk!
|
|
|
|
## Training and evaluation data
|
|
|
|
Fietje 2 instruct was finetuned from [the base model](https://huggingface.co/BramVanroy/fietje-2) on the following datasets. Number of training samples per dataset given in brackets, totalling 201,579 samples.
|
|
|
|
- [BramVanroy/ultrachat_200k_dutch](https://huggingface.co/datasets/BramVanroy/ultrachat_200k_dutch): gpt-4-1106-preview; multi-turn; fully generated (192,598)
|
|
- [BramVanroy/no_robots_dutch](https://huggingface.co/datasets/BramVanroy/no_robots_dutch): gpt-4-1106-preview; prompt translate, answer generated; some items have system messages (8181)
|
|
- [BramVanroy/belebele_dutch](https://huggingface.co/datasets/BramVanroy/belebele_dutch): Dutch portion of [belebele](https://huggingface.co/datasets/facebook/belebele), formatted into SFT format (800)
|
|
|
|
## Training procedure
|
|
|
|
I am thankful to the [Flemish Supercomputer Center](https://www.vscentrum.be/) (VSC) for providing the computational power to accomplish this project. Accounting for waiting for jobs, training took around a day on four nodes of 4x A100 80GB each (16 total). I cannot find the exact time anymore and I do not think that the runtime in `all_results.json` accounts for interrupted-and-continued jobs.
|
|
|
|
Training was done with the wonderful [alignment-handbook](https://github.com/huggingface/alignment-handbook), using DeepSpeed as a back-end. Exact training recipes and SLURM script are given in the [Github repository](https://github.com/BramVanroy/fietje).
|
|
|
|
|
|
### Training hyperparameters
|
|
|
|
The following hyperparameters were used during training:
|
|
- learning_rate: 6e-05
|
|
- train_batch_size: 42
|
|
- eval_batch_size: 42
|
|
- seed: 42
|
|
- distributed_type: multi-GPU
|
|
- num_devices: 16
|
|
- total_train_batch_size: 672
|
|
- total_eval_batch_size: 672
|
|
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-07
|
|
- lr_scheduler_type: cosine
|
|
- lr_scheduler_warmup_ratio: 0.1
|
|
- num_epochs: 3.0
|
|
|
|
### Training results
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss |
|
|
|:-------------:|:-----:|:----:|:---------------:|
|
|
| 0.9325 | 1.0 | 178 | 0.9060 |
|
|
| 0.8687 | 2.0 | 356 | 0.8850 |
|
|
| 0.8385 | 3.0 | 534 | 0.8818 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 4.39.1
|
|
- Pytorch 2.1.2+cu121
|
|
- Datasets 2.18.0
|
|
- Tokenizers 0.15.2
|