README.md

---
license: apache-2.0
datasets:
- CohereForAI/aya_dataset
- argilla/databricks-dolly-15k-curated-multilingual
- Gael540/dataSet_ens_sup_fr-v1
- ai2-adapt-dev/flan_v2_converted
- OpenAssistant/oasst1
language:
- fr
- en
- de
- it
- es
base_model:
- OpenLLM-France/Lucie-7B
pipeline_tag: text-generation
---

# Model Card for Lucie-7B-Instruct-human-data

* [Model Description](#model-description)
<!-- * [Uses](#uses) -->
* [Training Details](#training-details)
  * [Training Data](#training-data)
  * [Preprocessing](#preprocessing)
  * [Training Procedure](#training-procedure)
<!-- * [Evaluation](#evaluation) -->
* [Testing the model](#testing-the-model)
  * [Test with ollama](#test-with-ollama)
  * [Test with vLLM](#test-with-vllm)
* [Citation](#citation)
* [Acknowledgements](#acknowledgements)
* [Contact](#contact)

## Model Description

Lucie-7B-Instruct-human-data is a fine-tuned version of [Lucie-7B](), an open-source, multilingual causal language model created by OpenLLM-France.

Lucie-7B-Instruct-human-data is fine-tuned on human-produced instructions collected either from open annotation campaigns or by applying templates to extant datasets. The performance of Lucie-7B-Instruct-human-data falls below that of [Lucie-7B-Instruct](https://huggingface.co/OpenLLM-France/Lucie-7B-Instruct); the interest of the model is to show what can be done to fine-tune LLMs to follow instructions without appealing to third party LLMs.


## Training details
### Training data

Lucie-7B-Instruct-human-data is trained on the following datasets published by third parties:
* [Aya Dataset](https://huggingface.co/datasets/CohereForAI/aya_dataset) (English, 3944 samples; French, 1422; German, 241; Italian, 738; Spanish, 3854)
* [Dolly](https://huggingface.co/datasets/argilla/databricks-dolly-15k-curated-multilingual) (English, French, German, Spanish; 15015 x 4 samples)
* [ENS](https://huggingface.co/datasets/Gael540/dataSet_ens_sup_fr-v1) (French, 394 samples)
* [FLAN v2 Converted](https://huggingface.co/datasets/ai2-adapt-dev/flan_v2_converted) (English, 78580 samples)
* [Open Assistant 1](https://huggingface.co/datasets/OpenAssistant/oasst1) (English, 21151 samples; French, 1223; German, 1515; Italian, 370; Spanish, 14078)
* [Oracle](https://github.com/opinionscience/InstructionFr/tree/main/wikipedia) (French, 4613 samples)
* [PIAF](https://www.data.gouv.fr/fr/datasets/piaf-le-dataset-francophone-de-questions-reponses/) (French, 1849 samples)


And the following datasets developed for the Lucie instruct models:
* Croissant Aligned Instruct (French-English, 20K examples sampled randomly from 80K total)
* Hard-coded prompts concerning OpenLLM and Lucie (based on [allenai/tulu-3-hard-coded-10x](https://huggingface.co/datasets/allenai/tulu-3-hard-coded-10x))
    * French: openllm_french.jsonl (24x10 samples)
    * English: openllm_english.jsonl (24x10 samples)

### Preprocessing
* Filtering by language: Aya Dataset, Dolly and Open Assistant were filtered to keep only English and French samples, respectively.
* Filtering by keyword: Examples containing assistant responses were filtered out from Open Assistant if the responses contained a keyword from the list [filter_strings](https://github.com/OpenLLM-France/Lucie-Training/blob/98792a1a9015dcf613ff951b1ce6145ca8ecb174/tokenization/data.py#L2012). This filter is designed to remove examples in which the assistant is presented as model other than Lucie (e.g., ChatGPT, Gemma, Llama, ...).
* Duplicate examples were removed from Open Assistant.

### Training procedure

The model architecture and hyperparameters are the same as for [Lucie-7B](https://huggingface.co/OpenLLM-France/Lucie-7B) during the annealing phase with the following exceptions:
* context length: 4096 
* batch size: 1024
* max learning rate: 3e-5
* min learning rate: 3e-6


## Testing the model

### Test with ollama

* Download and install [Ollama](https://ollama.com/download)
* Download the [GGUF model](https://huggingface.co/OpenLLM-France/Lucie-7B-Instruct-v1/resolve/main/Lucie-7B-q4_k_m.gguf)
* Copy the [`Modelfile`](Modelfile), adpating if necessary the path to the GGUF file (line starting with `FROM`).
* Run in a shell:
    * `ollama create -f Modelfile Lucie`
    * `ollama run Lucie`
* Once ">>>" appears, type your prompt(s) and press Enter.
* Optionally, restart a conversation by typing "`/clear`"
* End the session by typing "`/bye`".

Useful for debug:
* [How to print input requests and output responses in Ollama server?](https://stackoverflow.com/a/78831840)
* [Documentation on Modelfile](https://github.com/ollama/ollama/blob/main/docs/modelfile.md#parameter)
   * Examples: [Ollama model library](https://github.com/ollama/ollama#model-library)
      * Llama 3 example: https://ollama.com/library/llama3.1
* Add GUI : https://docs.openwebui.com/

### Test with vLLM

#### 1. Run vLLM Docker Container

Use the following command to deploy the model,
replacing `INSERT_YOUR_HF_TOKEN` with your Hugging Face Hub token.

```bash
docker run --runtime nvidia --gpus=all \
    --env "HUGGING_FACE_HUB_TOKEN=INSERT_YOUR_HF_TOKEN" \
    -p 8000:8000 \
    --ipc=host \
    vllm/vllm-openai:latest \
    --model OpenLLM-France/Lucie-7B-Instruct-v1
```

#### 2. Test using OpenAI Client in Python

To test the deployed model, use the OpenAI Python client as follows:

```python
from openai import OpenAI

# Initialize the client
client = OpenAI(base_url='http://localhost:8000/v1', api_key='empty')

# Define the input content
content = "Hello Lucie"

# Generate a response
chat_response = client.chat.completions.create(
    model="OpenLLM-France/Lucie-7B-Instruct-v1",
    messages=[
        {"role": "user", "content": content}
    ],
)
print(chat_response.choices[0].message.content)
```

## Citation

Coming soon.

## Acknowledgements

This work was performed using HPC resources from GENCI–IDRIS (Grant 2024-GC011015444).

Lucie-7B was created by members of [LINAGORA](https://labs.linagora.com/) and the [OpenLLM-France](https://www.openllm-france.fr/) community, including in alphabetical order:
Olivier Gouvert (LINAGORA),
Ismaïl Harrando (LINAGORA/SciencesPo), 
Julie Hunter (LINAGORA),
Jean-Pierre Lorré (LINAGORA),
Jérôme Louradour (LINAGORA),
Michel-Marie Maudet (LINAGORA), and
Laura Rivière (LINAGORA).


We thank 
Clément Bénesse (Opsci), 
Christophe Cerisara (LORIA),
Evan Dufraisse (CEA),
Guokan Shang (MBZUAI), 
Joël Gombin (Opsci), 
Jordan Ricker (Opsci), 
and
Olivier Ferret (CEA) 
for their helpful input.

## Contact

[email protected]