Its currently refusing to answer things that LLaMa 3.1 405B has no issue answering. Deepseek is super censored at the moment, making it borderline unusable.
I saw these resources online that could help with liberating it:
https://gist.github.com/ruvnet/a4beba51960f6027edc003e05f3a350e
https://privatellm.app/blog/uncensored-deepseek-r1-distill-now-in-private-llm-for-ios-and-macos
https://ollama.com/huihui_ai/deepseek-r1-abliterated
https://huggingface.co/huihui-ai/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2
Please authenticate to join the conversation.
Completed
Feature Requests
About 1 year ago

ReactorcoreGames
Get notified by email when there are changes.
Completed
Feature Requests
About 1 year ago

ReactorcoreGames
Get notified by email when there are changes.