Update README.md
Browse files
README.md
CHANGED
|
@@ -13,6 +13,23 @@ base_model:
|
|
| 13 |
pipeline_tag: text-generation
|
| 14 |
---
|
| 15 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
## Model Description
|
| 17 |
|
| 18 |
Lucie-7B-Instruct is a fine-tuned version of [Lucie-7B](), an open-source, multilingual causal language model created by OpenLLM-France.
|
|
@@ -21,7 +38,6 @@ Lucie-7B-Instruct is fine-tuned on synthetic instructions produced by ChatGPT an
|
|
| 21 |
|
| 22 |
|
| 23 |
|
| 24 |
-
|
| 25 |
## Training details
|
| 26 |
|
| 27 |
### Training data
|
|
@@ -114,3 +130,36 @@ chat_response = client.chat.completions.create(
|
|
| 114 |
)
|
| 115 |
print(chat_response.choices[0].message.content)
|
| 116 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
pipeline_tag: text-generation
|
| 14 |
---
|
| 15 |
|
| 16 |
+
# Model Card for Lucie-7B-Instruct
|
| 17 |
+
|
| 18 |
+
* [Model Description](#model-description)
|
| 19 |
+
<!-- * [Uses](#uses) -->
|
| 20 |
+
* [Training Details](#training-details)
|
| 21 |
+
* [Training Data](#training-data)
|
| 22 |
+
* [Preprocessing](#preprocessing)
|
| 23 |
+
* [Training Procedure](#training-procedure)
|
| 24 |
+
<!-- * [Evaluation](#evaluation) -->
|
| 25 |
+
* [Testing the model](#testing-the-model)
|
| 26 |
+
* [Test in python](#test-in-python)
|
| 27 |
+
* [Test with ollama](#test-with-ollama)
|
| 28 |
+
* [Test with vLLM](#test-with-vllm)
|
| 29 |
+
* [Citation](#citation)
|
| 30 |
+
* [Acknowledgements](#acknowledgements)
|
| 31 |
+
* [Contact](#contact)
|
| 32 |
+
|
| 33 |
## Model Description
|
| 34 |
|
| 35 |
Lucie-7B-Instruct is a fine-tuned version of [Lucie-7B](), an open-source, multilingual causal language model created by OpenLLM-France.
|
|
|
|
| 38 |
|
| 39 |
|
| 40 |
|
|
|
|
| 41 |
## Training details
|
| 42 |
|
| 43 |
### Training data
|
|
|
|
| 130 |
)
|
| 131 |
print(chat_response.choices[0].message.content)
|
| 132 |
```
|
| 133 |
+
|
| 134 |
+
## Citation
|
| 135 |
+
|
| 136 |
+
Coming soon.
|
| 137 |
+
|
| 138 |
+
## Acknowledgements
|
| 139 |
+
|
| 140 |
+
This work was performed using HPC resources from GENCI–IDRIS (Grant 2024-GC011015444).
|
| 141 |
+
|
| 142 |
+
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:
|
| 143 |
+
Olivier Gouvert (LINAGORA),
|
| 144 |
+
Ismaïl Harrando (LINAGORA/SciencesPo),
|
| 145 |
+
Julie Hunter (LINAGORA),
|
| 146 |
+
Jean-Pierre Lorré (LINAGORA),
|
| 147 |
+
Jérôme Louradour (LINAGORA),
|
| 148 |
+
Michel-Marie Maudet (LINAGORA), and
|
| 149 |
+
Laura Rivière (LINAGORA).
|
| 150 |
+
|
| 151 |
+
|
| 152 |
+
We thank
|
| 153 |
+
Clément Bénesse (Opsci),
|
| 154 |
+
Christophe Cerisara (LORIA),
|
| 155 |
+
Evan Dufraisse (CEA),
|
| 156 |
+
Guokan Shang (MBZUAI),
|
| 157 |
+
Joël Gombin (Opsci),
|
| 158 |
+
Jordan Ricker (Opsci),
|
| 159 |
+
and
|
| 160 |
+
Olivier Ferret (CEA)
|
| 161 |
+
for their helpful input.
|
| 162 |
+
|
| 163 |
+
## Contact
|
| 164 |
+
|
| 165 |