metadata
base_model:
- Qwen/Qwen2.5-14B-Instruct-1M
A simple unalignment fine-tune on ~900k tokens aiming to make the model more compliant and willing to handle user requests.
This is the same unalignment training seen in concedo/Beepo-22B, so big thanks to concedo for the dataset.
Chat template is same as the original, ChatML.