Hugging Face's logo

TeeZee
/
Bielik-SOLAR-LIKE-10.7B-Instruct-v0.1

Text Generation
Transformers
Safetensors
mistral
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
Bielik-SOLAR-LIKE-10.7B-Instruct-v0.1 / README.md
TeeZee's picture
TeeZee
Update README.md
c7a9640 verified
preview code
|
raw
history blame contribute delete
1.09 kB
metadata
license: cc-by-nc-4.0

TeeZee/Bielik-SOLAR-LIKE-10.7B-Instruct-v0.1

Precise recipe used by Upstage to create SOLAR was applied to https://huggingface.co/speakleash/Bielik-7B-Instruct-v0.1 *(just merge, no finetuning)

Results

  • model is still coherent in Polish language, even without finetuning after merge

  • instruct mode works in ooba without issues

  • model is censored and aligned

  • seems that this model scores highest amongst all versions of original Bielik models, further finetunig should improve results even more.

    imgage/png

  • on dedicated to Polish speaking LLM leaderboards, its 2nd, just behind instruct version used for this merge, and thats to be expected when applying DUS merge - very small quality loss.

Polish LLMs leaderboards

  • overall it seems like a good base for further finetunig in Polish language.