README.md

---
language:
- nl
license: cc-by-nc-4.0
library_name: transformers
tags:
- trl
- dpo
- conversational
datasets:
- BramVanroy/ultra_feedback_dutch_cleaned
pipeline_tag: text-generation
inference: false
model-index:
- name: Qwen1.5-7B-Dutch-Chat
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: AI2 Reasoning Challenge (25-Shot)
      type: ai2_arc
      config: ARC-Challenge
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: acc_norm
      value: 53.92
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=robinsmits/Qwen1.5-7B-Dutch-Chat
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: HellaSwag (10-Shot)
      type: hellaswag
      split: validation
      args:
        num_few_shot: 10
    metrics:
    - type: acc_norm
      value: 76.03
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=robinsmits/Qwen1.5-7B-Dutch-Chat
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MMLU (5-Shot)
      type: cais/mmlu
      config: all
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 62.38
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=robinsmits/Qwen1.5-7B-Dutch-Chat
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: TruthfulQA (0-shot)
      type: truthful_qa
      config: multiple_choice
      split: validation
      args:
        num_few_shot: 0
    metrics:
    - type: mc2
      value: 45.34
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=robinsmits/Qwen1.5-7B-Dutch-Chat
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: Winogrande (5-shot)
      type: winogrande
      config: winogrande_xl
      split: validation
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 68.82
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=robinsmits/Qwen1.5-7B-Dutch-Chat
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: GSM8k (5-shot)
      type: gsm8k
      config: main
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 15.47
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=robinsmits/Qwen1.5-7B-Dutch-Chat
      name: Open LLM Leaderboard
---

# Qwen1.5-7B-Dutch-Chat

## Model description

This DPO aligned model is the merged version of the adapter model [robinsmits/Qwen1.5-7B-Dutch-Chat-Dpo](https://huggingface.co/robinsmits/Qwen1.5-7B-Dutch-Chat-Dpo). 

DPO Finetuning was performed on the Dutch [BramVanroy/ultra_feedback_dutch_cleaned](https://huggingface.co/datasets/BramVanroy/ultra_feedback_dutch_cleaned) dataset.

See [Qwen/Qwen1.5-7B-Chat](https://huggingface.co/Qwen/Qwen1.5-7B-Chat) for all information about the base model.



## ScandEval Dutch Leaderboard Evaluation Results

For evaluation results based on the Dutch language you can take a look at the site of ScandEval.

This model achieves a score which is very close to the performance of GPT-3.5.

[Dutch Natural Language Understanding](https://scandeval.com/dutch-nlu/)

[Dutch Natural Language Generation](https://scandeval.com/dutch-nlg/)



## [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_robinsmits__Qwen1.5-7B-Dutch-Chat)

Note that these Evaluation Results are for the English language.

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |53.66|
|AI2 Reasoning Challenge (25-Shot)|53.92|
|HellaSwag (10-Shot)              |76.03|
|MMLU (5-Shot)                    |62.38|
|TruthfulQA (0-shot)              |45.34|
|Winogrande (5-shot)              |68.82|
|GSM8k (5-shot)                   |15.47|


## Model usage

A basic example of how to use the finetuned model.

```
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

device = 'cuda'
model_name = 'robinsmits/Qwen1.5-7B-Dutch-Chat'

model = AutoModelForCausalLM.from_pretrained(model_name, 
                                             device_map = "auto", 
                                             torch_dtype = torch.bfloat16)

tokenizer = AutoTokenizer.from_pretrained(model_name)

messages = [{"role": "user", "content": "Hoi hoe gaat het ermee? Wat kun je me vertellen over appels?"}]

encoded_ids = tokenizer.apply_chat_template(messages, 
                                            add_generation_prompt = True,
                                            return_tensors = "pt")

generated_ids = model.generate(input_ids = encoded_ids.to(device), 
                               max_new_tokens = 256,
                               do_sample = True)
decoded = tokenizer.batch_decode(generated_ids)
print(decoded[0])
```

Below the chat template with the generated output.

```
<|im_start|>system
Je bent een behulpzame AI assistent<|im_end|>
<|im_start|>user
Hoi hoe gaat het ermee? Wat kun je me vertellen over appels?<|im_end|>
<|im_start|>assistant
Hallo! Appels zijn zo'n lekkere fruitsoort. Ze zijn zoet en knapperig, en je kunt ze koken, roosteren of zelfs in smoothies doen. Er zijn heel veel verschillende soorten appels, zoals de Fuji, Granny Smith en Gala. De appels die je meestal in de winkel koopt, komen van bomen die in het oosten van Noord-Amerika groeien.<|im_end|>
```

## Intended uses & limitations

As with all LLM's this model can also experience bias and hallucinations. Regardless of how you use this model always perform the necessary testing and validation.

The used dataset does not allow commercial usage.

## Training and evaluation data

The training notebook is available at the following link: [Qwen1_5_7B_Dutch_Chat_DPO](https://github.com/RobinSmits/Dutch-LLMs/blob/main/Qwen1_5_7B_Dutch_Chat_DPO.ipynb)

Training was performed with Google Colab PRO on a A100 - 40GB and lasted around 4 hours.

It achieves the following results on the evaluation set:
- Loss: 0.2610
- Rewards/chosen: -0.7248
- Rewards/rejected: -2.6224
- Rewards/accuracies: 0.9170
- Rewards/margins: 1.8976
- Logps/rejected: -877.8102
- Logps/chosen: -783.4282
- Logits/rejected: -0.8110
- Logits/chosen: -0.7528

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 1
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 32
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.05
- 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 |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
| 0.5503        | 0.1   | 30   | 0.4684          | -0.0439        | -0.6295          | 0.8919             | 0.5856          | -837.9513      | -769.8103    | -0.9335         | -0.8894       |
| 0.4178        | 0.2   | 60   | 0.3568          | -0.3713        | -1.4769          | 0.9015             | 1.1056          | -854.9000      | -776.3594    | -0.8768         | -0.8276       |
| 0.3264        | 0.29  | 90   | 0.3143          | -0.4893        | -1.8730          | 0.9151             | 1.3837          | -862.8228      | -778.7191    | -0.8428         | -0.7929       |
| 0.2999        | 0.39  | 120  | 0.2885          | -0.6832        | -2.3118          | 0.9151             | 1.6286          | -871.5981      | -782.5971    | -0.8260         | -0.7730       |
| 0.3454        | 0.49  | 150  | 0.2749          | -0.7239        | -2.4904          | 0.9189             | 1.7664          | -875.1693      | -783.4113    | -0.8235         | -0.7678       |
| 0.3354        | 0.59  | 180  | 0.2685          | -0.6775        | -2.4859          | 0.9170             | 1.8084          | -875.0795      | -782.4824    | -0.8130         | -0.7574       |
| 0.2848        | 0.68  | 210  | 0.2652          | -0.7157        | -2.5692          | 0.9131             | 1.8535          | -876.7465      | -783.2466    | -0.8157         | -0.7586       |
| 0.3437        | 0.78  | 240  | 0.2621          | -0.7233        | -2.6091          | 0.9151             | 1.8857          | -877.5430      | -783.3994    | -0.8138         | -0.7561       |
| 0.2655        | 0.88  | 270  | 0.2611          | -0.7183        | -2.6154          | 0.9151             | 1.8971          | -877.6708      | -783.2995    | -0.8106         | -0.7524       |
| 0.3442        | 0.98  | 300  | 0.2610          | -0.7248        | -2.6224          | 0.9170             | 1.8976          | -877.8102      | -783.4282    | -0.8110         | -0.7528       |

### Framework versions

- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2

## Citation
Thanks to the creators of Qwen1.5 for their great work!
```
@article{qwen,
  title={Qwen Technical Report},
  author={Jinze Bai and Shuai Bai and Yunfei Chu and Zeyu Cui and Kai Dang and Xiaodong Deng and Yang Fan and Wenbin Ge and Yu Han and Fei Huang and Binyuan Hui and Luo Ji and Mei Li and Junyang Lin and Runji Lin and Dayiheng Liu and Gao Liu and Chengqiang Lu and Keming Lu and Jianxin Ma and Rui Men and Xingzhang Ren and Xuancheng Ren and Chuanqi Tan and Sinan Tan and Jianhong Tu and Peng Wang and Shijie Wang and Wei Wang and Shengguang Wu and Benfeng Xu and Jin Xu and An Yang and Hao Yang and Jian Yang and Shusheng Yang and Yang Yao and Bowen Yu and Hongyi Yuan and Zheng Yuan and Jianwei Zhang and Xingxuan Zhang and Yichang Zhang and Zhenru Zhang and Chang Zhou and Jingren Zhou and Xiaohuan Zhou and Tianhang Zhu},
  journal={arXiv preprint arXiv:2309.16609},
  year={2023}
}
```