|
--- |
|
license: cc-by-nc-4.0 |
|
base_model: snoels/FinGEITje-7B-sft |
|
datasets: |
|
- BramVanroy/ultra_feedback_dutch |
|
library_name: peft |
|
tags: |
|
- alignment-handbook |
|
- trl |
|
- dpo |
|
- generated_from_trainer |
|
- geitje |
|
- fingeitje |
|
- dutch |
|
- nl |
|
- finance |
|
model-index: |
|
- name: snoels/FinGEITje-7B-dpo |
|
results: [] |
|
language: |
|
- nl |
|
pipeline_tag: text-generation |
|
inference: false |
|
--- |
|
|
|
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/snoels/huggingface/runs/yng7mdb0) |
|
|
|
<p align="center" style="margin:0;padding:0"> |
|
<img src="https://huggingface.co/snoels/FinGEITje-7B-dpo/resolve/main/fingeitje-banner-dpo.png" alt="FinGEITje DPO Banner" width="1000"/> |
|
</p> |
|
|
|
<div style="margin:auto; text-align:center"> |
|
<h1 style="margin-bottom: 0; font-size: 2em;">🐐 FinGEITje 7B DPO</h1> |
|
<em style="font-size: 1em;">A large open Dutch financial language model aligned through AI feedback.</em> |
|
</div> |
|
|
|
This model is a fine-tuned version of [snoels/FinGEITje-7B-sft](https://huggingface.co/snoels/FinGEITje-7B-sft) on the [BramVanroy/ultra_feedback_dutch](https://huggingface.co/datasets/BramVanroy/ultra_feedback_dutch) dataset. |
|
|
|
## 📖 Model Description |
|
|
|
[FinGEITje-7B-dpo](https://huggingface.co/snoels/FinGEITje-7B-dpo) is a large open Dutch financial language model with 7 billion parameters, based on Mistral 7B. It has been further trained using **Direct Preference Optimization (DPO)** on AI-generated preference data, aligning the model's responses with human-like preferences in the Dutch language. This alignment process enhances the model's ability to generate more helpful, coherent, and user-aligned responses in financial contexts. |
|
|
|
## 📊 Training |
|
|
|
### Training Data |
|
|
|
[FinGEITje-7B-dpo](https://huggingface.co/snoels/FinGEITje-7B-dpo) was fine-tuned on the [BramVanroy/ultra_feedback_dutch](https://huggingface.co/datasets/BramVanroy/ultra_feedback_dutch) dataset, which consists of synthetic preference data in Dutch. This dataset includes prompts along with preferred and less preferred responses, allowing the model to learn to generate more aligned responses through DPO. |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 5e-06 |
|
- train_batch_size: 1 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 4 |
|
- gradient_accumulation_steps: 16 |
|
- total_train_batch_size: 64 |
|
- total_eval_batch_size: 4 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen | |
|
|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:| |
|
| 0.1029 | 0.1327 | 100 | 0.1099 | -1.8067 | -5.3683 | 0.9679 | 3.5616 | -892.3373 | -579.9115 | -2.4775 | -2.3705 | |
|
| 0.042 | 0.2654 | 200 | 0.0430 | -3.5129 | -10.6778 | 0.9828 | 7.1649 | -1423.2883 | -750.5289 | -1.9744 | -1.9895 | |
|
| 0.0278 | 0.3981 | 300 | 0.0344 | -3.7335 | -13.5153 | 0.9828 | 9.7818 | -1707.0360 | -772.5893 | -1.7454 | -1.8191 | |
|
| 0.0223 | 0.5308 | 400 | 0.0308 | -3.6554 | -13.7712 | 0.9858 | 10.1158 | -1732.6289 | -764.7831 | -1.8020 | -1.9184 | |
|
| 0.0378 | 0.6635 | 500 | 0.0297 | -4.0018 | -16.3285 | 0.9851 | 12.3266 | -1988.3542 | -799.4221 | -1.6924 | -1.8650 | |
|
| 0.0352 | 0.7962 | 600 | 0.0278 | -3.8104 | -15.6430 | 0.9836 | 11.8327 | -1919.8119 | -780.2752 | -1.7437 | -1.8978 | |
|
| 0.0238 | 0.9289 | 700 | 0.0279 | -3.8974 | -15.9642 | 0.9828 | 12.0668 | -1951.9310 | -788.9780 | -1.7371 | -1.8937 | |
|
|
|
### Framework versions |
|
|
|
- PEFT 0.11.1 |
|
- Transformers 4.42.4 |
|
- Pytorch 2.3.1 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
|
|
|
## 🛠️ How to Use |
|
|
|
[FinGEITje-7B-dpo](https://huggingface.co/snoels/FinGEITje-7B-dpo) can be utilized using the Hugging Face Transformers library along with PEFT to load the adapters efficiently. |
|
|
|
### Installation |
|
|
|
Ensure you have the necessary libraries installed: |
|
|
|
```bash |
|
pip install torch transformers peft accelerate |
|
``` |
|
|
|
### Loading the Model |
|
|
|
```python |
|
from transformers import AutoTokenizer, AutoModelForCausalLM |
|
from peft import PeftModel |
|
|
|
# Load the tokenizer |
|
tokenizer = AutoTokenizer.from_pretrained("BramVanroy/GEITje-7B-ultra", use_fast=False) |
|
|
|
# Load the base model |
|
base_model = AutoModelForCausalLM.from_pretrained("BramVanroy/GEITje-7B-ultra", device_map='auto') |
|
|
|
# Load the FinGEITje-7B-dpo model with PEFT adapters |
|
model = PeftModel.from_pretrained(base_model, "snoels/FinGEITje-7B-dpo", device_map='auto') |
|
``` |
|
|
|
### Generating Text |
|
|
|
```python |
|
# Prepare the input |
|
input_text = "Wat zijn de laatste trends in de Nederlandse banksector?" |
|
input_ids = tokenizer.encode(input_text, return_tensors='pt').to(model.device) |
|
|
|
# Generate a response |
|
outputs = model.generate(input_ids, max_length=200, num_return_sequences=1) |
|
response = tokenizer.decode(outputs[0], skip_special_tokens=True) |
|
|
|
print(response) |
|
``` |
|
|
|
## 🙏 Acknowledgements |
|
|
|
We would like to thank: |
|
|
|
- **Rijgersberg** ([GitHub](https://github.com/Rijgersberg)) for creating [GEITje](https://github.com/Rijgersberg/GEITje), one of the first Dutch foundation models. |
|
- **Bram Vanroy** ([GitHub](https://github.com/BramVanroy)) for creating [GEITje-7B-ultra](https://huggingface.co/BramVanroy/GEITje-7B-ultra) and providing the ultra_feedback_dutch dataset. |
|
- **Contributors of the [Alignment Handbook](https://github.com/huggingface/alignment-handbook)** for providing valuable resources that guided the development and training process of [FinGEITje-7B-dpo](https://huggingface.co/snoels/FinGEITje-7B-dpo). |
|
- **Silverfin** for their collaboration in this research. Silverfin, a Belgian scale-up focused on building an accountancy cloud service, provided valuable insights and resources that were instrumental in the development of FinGEITje. More about their work can be found at [Silverfin](https://silverfin.com/). |
|
|
|
## 📝 Citation |
|
[Link to the paper](https://arxiv.org/abs/2410.12835) |
|
|
|
If you use [FinGEITje-7B-dpo](https://huggingface.co/snoels/FinGEITje-7B-dpo) in your work, please cite: |
|
|
|
```bibtex |
|
@article{FinGEITje2024, |
|
title={A Dutch Financial Large Language Model}, |
|
author={Noels, Sander and De Blaere, Jorne and De Bie, Tijl}, |
|
journal={arXiv preprint arXiv:2410.12835}, |
|
year={2024}, |
|
url={https://arxiv.org/abs/2410.12835} |
|
} |
|
``` |
|
|
|
## 📜 License |
|
|
|
This model is licensed under the [Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)](https://creativecommons.org/licenses/by-nc/4.0/) license. |
|
|
|
## 📧 Contact |
|
|
|
For any inquiries or questions, please contact [Sander Noels](mailto:[email protected]). |
