|
--- |
|
language: |
|
- en |
|
- de |
|
library_name: transformers |
|
pipeline_tag: text-generation |
|
license: apache-2.0 |
|
--- |
|
|
|
 |
|
|
|
_Hermes + Leo = Hermeo_ |
|
|
|
# Hermeo-7B |
|
|
|
A German-English language model merged from [DPOpenHermes-7B-v2](https://huggingface.co/openaccess-ai-collective/DPOpenHermes-7B-v2) and [leo-mistral-hessianai-7b-chat](https://huggingface.co/LeoLM/leo-mistral-hessianai-7b-chat) using [mergekit](https://github.com/cg123/mergekit). |
|
Both base models are fine-tuned versions of [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1). |
|
|
|
|
|
### Model details |
|
|
|
- **Merged from:** [leo-mistral-hessianai-7b-chat](https://huggingface.co/LeoLM/leo-mistral-hessianai-7b-chat) and [DPOpenHermes-7B-v2](https://huggingface.co/openaccess-ai-collective/DPOpenHermes-7B-v2) |
|
- **Model type:** Causal decoder-only transformer language model |
|
- **Languages:** English and German |
|
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0.html) |
|
|
|
### Acknowledgements |
|
|
|
- This model release is heavily inspired by [Weyaxi/OpenHermes-2.5-neural-chat-v3-2-Slerp](https://huggingface.co/Weyaxi/OpenHermes-2.5-neural-chat-v3-2-Slerp) |
|
- Thanks to the authors of the base models: [Mistral](https://mistral.ai/), [LAION](https://laion.ai/), [HessianAI](https://hessian.ai/), [Open Access AI Collective](https://huggingface.co/openaccess-ai-collective), [@teknium](https://huggingface.co/teknium), [@bjoernp](https://huggingface.co/bjoernp) |
|
- The [German evaluation datasets and scripts](https://github.com/bjoernpl/lm-evaluation-harness-de/tree/mmlu_de) from [@bjoernp](https://huggingface.co/bjoernp) were used. |
|
- The computing resources from [DFKI's PEGASUS cluster](https://pegasus.dfki.de/) were used for the evaluation. |
|
|
|
|
|
## Evaluation |
|
|
|
The evaluation methdology of the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) is followed. |
|
|
|
### German benchmarks |
|
|
|
| **German tasks:** | **MMLU-DE** | **Hellaswag-DE** | **ARC-DE** | |
|
|-------------------------------|-------------|---------------|--------------| |
|
| **Models / Few-shots:** | _(5 shots)_ | _(10 shots)_ | _(24 shots)_ | |
|
| _7B parameters_ | | | | |
|
| llama-2-7b | 0.400 | 0.513 | 0.381 | |
|
| leo-hessianai-7b | 0.400 | 0.609 | 0.429 | |
|
| bloom-6b4-clp-german | 0.274 | 0.550 | 0.351 | |
|
| mistral-7b | **0.524** | 0.588 | 0.473 | |
|
| leo-mistral-hessianai-7b | 0.481 | 0.663 | 0.485 | |
|
| leo-mistral-hessianai-7b-chat | 0.458 | 0.617 | 0.465 | |
|
| DPOpenHermes-7B-v2 | TBA | 0.603 | 0.515 | |
|
| hermeo-7b (this model) | 0.511 | **0.668** | **0.528** | |
|
| _13B parameters_ | | | | |
|
| llama-2-13b | 0.469 | 0.581 | 0.468 | |
|
| leo-hessianai-13b | **0.486** | **0.658** | **0.509** | |
|
| _70B parameters_ | | | | |
|
| llama-2-70b | 0.597 | 0.674 | 0.561 | |
|
| leo-hessianai-70b | **0.653** | **0.721** | **0.600** | |
|
|
|
### English benchmarks |
|
|
|
TBA |
|
|
|
## Prompting / Prompt Template |
|
|
|
Prompt dialogue template (ChatML format): |
|
|
|
``` |
|
""" |
|
<|im_start|>system |
|
{system_message}<|im_end|> |
|
<|im_start|>user |
|
{prompt}<|im_end|> |
|
<|im_start|>assistant |
|
""" |
|
``` |
|
|
|
The model input can contain multiple conversation turns between user and assistant, e.g. |
|
``` |
|
<|im_start|>user |
|
{prompt 1}<|im_end|> |
|
<|im_start|>assistant |
|
{reply 1}<|im_end|> |
|
<|im_start|>user |
|
{prompt 2}<|im_end|> |
|
<|im_start|>assistant |
|
(...) |
|
``` |
|
|
|
## License |
|
|
|
[Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0.html) |
|
|
