Create README.md

d43c215 verified 8 months ago

8.63 kB

	---
	license: apache-2.0
	language:
	- en
	- it
	pipeline_tag: text-generation
	---

	![image/png](https://huggingface.co./datasets/malteos/images/resolve/main/occiglot.medium.png)

	# Occiglot-7B-IT-EN

	> A [polyglot](https://en.wikipedia.org/wiki/Multilingualism#In_individuals) language model for the [Occident](https://en.wikipedia.org/wiki/Occident).
	>

	Occiglot-7B-IT-EN is a generative language model with 7B parameters for Italian and English and trained by the [Occiglot Research Collective](https://occiglot.github.io/occiglot/)..
	It is based on [Mistral-7B-v0.1](https://huggingface.co./mistralai/Mistral-7B-v0.1) and trained on 113B tokens of additional multilingual and code data with a block size of 8,192 tokens per sample.
	Note that the model is a general-purpose base model and was not instruction-fine-tuned nor optimized for chat or other applications. We make an instruction tuned variant available as [occiglot-7b-it-en-instruct](https://huggingface.co./occiglot/occiglot-7b-it-en-instruct)

	This is the first release of an ongoing open research project for multilingual language models.
	If you want to train a model for your own language or are working on evaluations, please contact us or join our [Discord server](https://discord.gg/wUpvYs4XvM). We are open for collaborations!


	### Model details

	- Continued-pretraining from: [Mistral-7B-v0.1](https://huggingface.co./mistralai/Mistral-7B-v0.1)
	- Model type: Causal decoder-only transformer language model
	- Languages: English, Italian, and code.
	- License: [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0.html)
	- Compute resources: [HessianAI's 42](https://hessian.ai/)
	- Contributors: Manuel Brack, Patrick Schramowski, Pedro Ortiz, Malte Ostendorff, Fabio Barth, Georg Rehm, Kristian Kersting
	- Research labs: [Occiglot](https://occiglot.github.io/occiglot/) with support from [SAINT](https://www.dfki.de/en/web/research/research-departments/foundations-of-systems-ai) and [SLT](https://www.dfki.de/en/web/research/research-departments/speech-and-language-technology)
	- Contact: [Discord](https://discord.gg/wUpvYs4XvM)

	### How to use

	You can use this model directly with a pipeline for text generation. Since the generation relies on some randomness, we
	set a seed for reproducibility:

	```python
	>>> from transformers import pipeline, set_seed
	>>> generator = pipeline('text-generation', model='occiglot/occiglot-7b-it-en')
	>>> set_seed(42)
	>>> generator("Salve, sono una modella linguistica,", max_length=40, num_return_sequences=1)
	[{'generated_text': 'Salve, sono una modella linguistica che può aiutarvi a tradurre testi tra l'italiano e l'inglese. Se mi inviate un testo in italiano'}]
	```

	## Dataset

	The training data is the respective subset of the data used for [occiglot-7b-eu5](https://huggingface.co./occiglot/occiglot-7b-eu5), i.e. Italian plus English and Code.

	The data distribution by language (estimated) is as follows:
	- English: ~34%
	- Code: ~13%
	- Italian: ~52%

	The training data was prepared using [lm-datasets](https://github.com/malteos/lm-datasets).
	The exact data configuration is [here](https://huggingface.co./occiglot/occiglot-7b-eu5/blob/main/lm-datasets-config.yml).

	## Training settings

	- Continual pre-training on 128 x A100-80GB on [HessianAI's 42](https://hessian.ai/).
	- Framework: [Determined](https://www.determined.ai/)
	- Precision: bf16
	- Optimizer: AdamW (lr: 0.00001, warmup_steps: 420)
	- Global batch size: 512 (with 8192 blocksize) split over 128 GPUs
	- Cosine Annealing with Warmup


	## Tokenizer

	Tokenizer is unchanged from [Mistral-7B-v0.1](https://huggingface.co./mistralai/Mistral-7B-v0.1).

	## Evaluation

	Preliminary evaluation results can be found below.
	Please note that the non-English results are based on partially machine-translated datasets and English prompts ([Belebele](https://huggingface.co./datasets/facebook/belebele) and [Okapi framework](https://github.com/nlp-uoregon/Okapi)) and thus should be interpreted with caution, e.g., biased towards English model performance.
	Currently, we are working on more suitable benchmarks for Spanish, French, German, and Italian.

	<details>
	<summary>Evaluation results</summary>

	### All 5 Languages

	\| \| avg \| arc_challenge \| belebele \| hellaswag \| mmlu \| truthfulqa \|
	\|:---------------------------\|---------:\|----------------:\|-----------:\|------------:\|---------:\|-------------:\|
	\| Occiglot-7b-eu5 \| 0.516895 \| 0.508109 \| 0.675556 \| 0.718963 \| 0.402064 \| 0.279782 \|
	\| Occiglot-7b-eu5-instruct \| 0.537799 \| 0.53632 \| 0.691111 \| 0.731918 \| 0.405198 \| 0.32445 \|
	\| Occiglot-7b-it-en \| 0.513221 \| 0.500564 \| 0.694444 \| 0.668099 \| 0.413528 \| 0.289469 \|
	\| Occiglot-7b-it-en-instruct \| 0.53721 \| 0.523128 \| 0.726667 \| 0.683414 \| 0.414918 \| 0.337927 \|
	\| Cerbero-7b \| 0.532385 \| 0.513714 \| 0.743111 \| 0.654061 \| 0.427566 \| 0.323475 \|
	\| Mistral-7b-v0.1 \| 0.547111 \| 0.528937 \| 0.768444 \| 0.682516 \| 0.448253 \| 0.307403 \|
	\| Mistral-7b-instruct-v0.2 \| 0.56713 \| 0.547228 \| 0.741111 \| 0.69455 \| 0.422501 \| 0.430262 \|


	### English

	\| \| avg \| arc_challenge \| belebele \| hellaswag \| mmlu \| truthfulqa \|
	\|:---------------------------\|---------:\|----------------:\|-----------:\|------------:\|---------:\|-------------:\|
	\| Occiglot-7b-eu5 \| 0.59657 \| 0.530717 \| 0.726667 \| 0.789882 \| 0.531904 \| 0.403678 \|
	\| Occiglot-7b-eu5-instruct \| 0.617905 \| 0.558874 \| 0.746667 \| 0.799841 \| 0.535109 \| 0.449 \|
	\| Occiglot-7b-it-en \| 0.630127 \| 0.580205 \| 0.774444 \| 0.804222 \| 0.578977 \| 0.412786 \|
	\| Occiglot-7b-it-en-instruct \| 0.659383 \| 0.609215 \| 0.82 \| 0.809301 \| 0.578835 \| 0.479562 \|
	\| Cerbero-7b \| 0.66661 \| 0.613481 \| 0.827778 \| 0.810396 \| 0.600484 \| 0.480911 \|
	\| Mistral-7b-v0.1 \| 0.668385 \| 0.612628 \| 0.844444 \| 0.834097 \| 0.624555 \| 0.426201 \|
	\| Mistral-7b-instruct-v0.2 \| 0.713657 \| 0.637372 \| 0.824444 \| 0.846345 \| 0.59201 \| 0.668116 \|

	### Italian

	\| \| avg \| arc_challenge_it \| belebele_it \| hellaswag_it \| mmlu_it \| truthfulqa_it \|
	\|:---------------------------\|---------:\|-------------------:\|--------------:\|---------------:\|----------:\|----------------:\|
	\| Occiglot-7b-eu5 \| 0.421382 \| 0.501283 \| 0.652222 \| 0.700533 \| 0 \| 0.252874 \|
	\| Occiglot-7b-eu5-instruct \| 0.437214 \| 0.516681 \| 0.661111 \| 0.71326 \| 0 \| 0.295019 \|
	\| Occiglot-7b-it-en \| 0.432667 \| 0.536356 \| 0.684444 \| 0.694768 \| 0 \| 0.247765 \|
	\| Occiglot-7b-it-en-instruct \| 0.456261 \| 0.545766 \| 0.717778 \| 0.713804 \| 0 \| 0.303959 \|
	\| Cerbero-7b \| 0.434939 \| 0.522669 \| 0.717778 \| 0.631567 \| 0 \| 0.302682 \|
	\| Mistral-7b-v0.1 \| 0.426264 \| 0.502139 \| 0.734444 \| 0.630371 \| 0 \| 0.264368 \|
	\| Mistral-7b-instruct-v0.2 \| 0.442383 \| 0.519247 \| 0.703333 \| 0.6394 \| 0 \| 0.349936 \|


	</details>

	## Acknowledgements

	The model training was supported by a compute grant at the [42 supercomputer](https://hessian.ai/) which is a central component in the development of [hessian AI](https://hessian.ai/), the [AI Innovation Lab](https://hessian.ai/infrastructure/ai-innovationlab/) (funded by the [Hessian Ministry of Higher Education, Research and the Art (HMWK)](https://wissenschaft.hessen.de) & the [Hessian Ministry of the Interior, for Security and Homeland Security (HMinD)](https://innen.hessen.de)) and the [AI Service Centers](https://hessian.ai/infrastructure/ai-service-centre/) (funded by the [German Federal Ministry for Economic Affairs and Climate Action (BMWK)](https://www.bmwk.de/Navigation/EN/Home/home.html)).
	The curation of the training data is partially funded by the [German Federal Ministry for Economic Affairs and Climate Action (BMWK)](https://www.bmwk.de/Navigation/EN/Home/home.html)
	through the project [OpenGPT-X](https://opengpt-x.de/en/) (project no. 68GX21007D).


	## License

	[Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0.html)

	## See also

	- https://huggingface.co./collections/occiglot/occiglot-eu5-7b-v01-65dbed502a6348b052695e01