|
--- |
|
inference: false |
|
library_name: transformers |
|
language: |
|
- en |
|
- fr |
|
- de |
|
- es |
|
- it |
|
- pt |
|
- ja |
|
- ko |
|
- zh |
|
- ar |
|
- el |
|
- fa |
|
- pl |
|
- id |
|
- cs |
|
- he |
|
- hi |
|
- nl |
|
- ro |
|
- ru |
|
- tr |
|
- uk |
|
- vi |
|
license: cc-by-nc-4.0 |
|
--- |
|
|
|
# Model Card for Aya Expanse 8B Ungated |
|
|
|
Aya-Expanse 8B, but not gated! |
|
|
|
<img src="aya-expanse-8B.png" width="650" style="margin-left:'auto' margin-right:'auto' display:'block'"/> |
|
|
|
**Aya Expanse 8B** is an open-weight research release of a model with highly advanced multilingual capabilities. It focuses on pairing a highly performant pre-trained [Command family](https://huggingface.co/CohereForAI/c4ai-command-r-plus) of models with the result of a year’s dedicated research from [Cohere For AI](https://cohere.for.ai/), including [data arbitrage](https://arxiv.org/abs/2408.14960), [multilingual preference training](https://arxiv.org/abs/2407.02552), [safety tuning](https://arxiv.org/abs/2406.18682), and [model merging](https://arxiv.org/abs/2410.10801). The result is a powerful multilingual large language model. |
|
|
|
This model card corresponds to the 8-billion version of the Aya Expanse model. We also released an 32-billion version which you can find [here](https://huggingface.co/CohereForAI/aya-expanse-32B). |
|
|
|
- Developed by: [Cohere For AI](https://cohere.for.ai/) |
|
- Point of Contact: Cohere For AI: [cohere.for.ai](https://cohere.for.ai/) |
|
- License: [CC-BY-NC](https://cohere.com/c4ai-cc-by-nc-license), requires also adhering to [C4AI's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy) |
|
- Model: Aya Expanse 8B |
|
- Model Size: 8 billion parameters |
|
|
|
### Supported Languages |
|
|
|
We cover 23 languages: Arabic, Chinese (simplified & traditional), Czech, Dutch, English, French, German, Greek, Hebrew, Hebrew, Hindi, Indonesian, Italian, Japanese, Korean, Persian, Polish, Portuguese, Romanian, Russian, Spanish, Turkish, Ukrainian, and Vietnamese. |
|
|
|
### Try it: Aya Expanse in Action |
|
|
|
Use the [Cohere playground](https://dashboard.cohere.com/playground/chat) or our [Hugging Face Space](https://huggingface.co/spaces/CohereForAI/aya_expanse) for interactive exploration. |
|
|
|
### How to Use Aya Expanse |
|
|
|
Install the transformers library and load Aya Expanse 8B as follows: |
|
|
|
```python |
|
from transformers import AutoTokenizer, AutoModelForCausalLM |
|
|
|
model_id = "CohereForAI/aya-expanse-8b" |
|
tokenizer = AutoTokenizer.from_pretrained(model_id) |
|
model = AutoModelForCausalLM.from_pretrained(model_id) |
|
|
|
# Format the message with the chat template |
|
messages = [{"role": "user", "content": "Anneme onu ne kadar sevdiğimi anlatan bir mektup yaz"}] |
|
input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt") |
|
## <BOS_TOKEN><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Anneme onu ne kadar sevdiğimi anlatan bir mektup yaz<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|> |
|
|
|
gen_tokens = model.generate( |
|
input_ids, |
|
max_new_tokens=100, |
|
do_sample=True, |
|
temperature=0.3, |
|
) |
|
|
|
gen_text = tokenizer.decode(gen_tokens[0]) |
|
print(gen_text) |
|
``` |
|
|
|
### Example Notebooks |
|
|
|
**Fine-Tuning:** |
|
- [Detailed Fine-Tuning Notebook](https://colab.research.google.com/drive/1ryPYXzqb7oIn2fchMLdCNSIH5KfyEtv4). |
|
|
|
**Community-Contributed Use Cases:**: |
|
|
|
The following notebooks contributed by *Cohere For AI Community* members show how Aya Expanse can be used for different use cases: |
|
- [Mulitlingual Writing Assistant](https://colab.research.google.com/drive/1SRLWQ0HdYN_NbRMVVUHTDXb-LSMZWF60) |
|
- [AyaMCooking](https://colab.research.google.com/drive/1-cnn4LXYoZ4ARBpnsjQM3sU7egOL_fLB?usp=sharing) |
|
- [Multilingual Question-Answering System](https://colab.research.google.com/drive/1bbB8hzyzCJbfMVjsZPeh4yNEALJFGNQy?usp=sharing) |
|
|
|
|
|
## Model Details |
|
|
|
**Input**: Models input text only. |
|
|
|
**Output**: Models generate text only. |
|
|
|
**Model Architecture**: Aya Expanse 8B is an auto-regressive language model that uses an optimized transformer architecture. Post-training includes supervised finetuning, preference training, and model merging. |
|
|
|
**Languages covered**: The model is particularly optimized for multilinguality and supports the following languages: Arabic, Chinese (simplified & traditional), Czech, Dutch, English, French, German, Greek, Hebrew, Hindi, Indonesian, Italian, Japanese, Korean, Persian, Polish, Portuguese, Romanian, Russian, Spanish, Turkish, Ukrainian, and Vietnamese |
|
|
|
**Context length**: 8K |
|
|
|
For more details about how the model was trained, check out [our blogpost](https://huggingface.co/blog/aya-expanse). |
|
|
|
### Evaluation |
|
|
|
We evaluated Aya Expanse 8B against Gemma 2 9B, Llama 3.1 8B, Ministral 8B, and Qwen 2.5 7B using Dolly and m-ArenaHard, a dataset based on the [Arena-Hard-Auto dataset](https://huggingface.co/datasets/lmarena-ai/arena-hard-auto-v0.1) and translated to the 23 languages we support in Aya Expanse 8B. Win-rates were determined using gpt-4o-2024-08-06 as a judge. For a conservative benchmark, we report results from gpt-4o-2024-08-06, though gpt-4o-mini scores showed even stronger performance. |
|
|
|
The m-ArenaHard dataset, used to evaluate Aya Expanse’s capabilities, is publicly available [here](https://huggingface.co/datasets/CohereForAI/m-ArenaHard). |
|
|
|
<img src="winrates_marenahard_complete.png" width="650" style="margin-left:'auto' margin-right:'auto' display:'block'"/> |
|
<img src="winrates_dolly.png" width="650" style="margin-left:'auto' margin-right:'auto' display:'block'"/> |
|
<img src="winrates_by_lang.png" width="650" style="margin-left:'auto' margin-right:'auto' display:'block'"/> |
|
<img src="winrates_step_by_step.png" width="650" style="margin-left:'auto' margin-right:'auto' display:'block'"/> |
|
|
|
|
|
### Model Card Contact |
|
|
|
For errors or additional questions about details in this model card, contact [email protected]. |
|
|
|
### Terms of Use |
|
|
|
We hope that the release of this model will make community-based research efforts more accessible, by releasing the weights of a highly performant multilingual model to researchers all over the world. This model is governed by a [CC-BY-NC](https://cohere.com/c4ai-cc-by-nc-license) License with an acceptable use addendum, and also requires adhering to [C4AI's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy). |
|
|