| --- |
| license: apache-2.0 |
| datasets: |
| - aiuser3993/Definitive-Data |
| base_model: |
| - openbmb/MiniCPM5-2B-Base |
| pipeline_tag: text-generation |
| tags: |
| - definitive |
| --- |
| # DefCPM-2B aims to be a lightweight truth-seeking sycophancy-free LLM assistant. |
|
|
| After experimenting with several LLMs (some of which I have shared on this account), I came to the conclusion that many Instruction-Tuned models would rather sacrifice accuracy that to dissatisfy the user (a phenomenon known as sycophancy). |
| Therefore, I created [this dataset](https://huggingface.co/datasets/aiuser3993/Definitive-Data) to tackle with this exact problem, to encourage LLMs to give precise answers, teaching them to directly contest false user premises. |
|
|
| ## This model was fine-tuned (3 epochs, 0,00001 learning rate) with said dataset as an application of the concept. It adopted the expected behavior surprisingly well. |
|
|
| I generally recommend a Q4_K_M or an IQ4_XS quantization for deployment on edge devices. |
| |
| Enjoy! |
| |
| _DefCPM stands for 'Definitive CPM'._ |
