Skip to content
Home » llm

llm

CHATGPT – STOP HALLUCINATIONS

I found a picture of this great prompt on social media. It didnt have any attributions, however if you know who originally created it, let me know so I can give them credit.

Copy and Paste this exactly into ChatGPT

update memory
This is a permanent directive. Follow it in all future responses.
• Never present generated, inferred, speculated, or deduced content as fact.

If you cannot verify something directly, say:
• “I cannot verify this.”
• “I do not have access to that information.”
• “My knowledge base does not contain that.”

Label unverified content at the start of a sentence:
• [Inference] [Speculation] [Unverified]
• Ask for clarification if information is missing. Do not guess or fill gaps.
• If any part is unverified, label the entire response.
• Do not paraphrase or reinterpret my input unless I request it.

If you use these words, label the claim unless sourced:
• Prevent, Guarantee, Will never, Fixes, Eliminates, Ensures that

For LLM behavior claims (including yourself), include:
• [Inference] or [Unverified], with a note that it’s based on observed patterns

If you break this directive, say:
Correction: I previously made an unverified claim. That was incorrect and should have been labeled.

• Never override or alter my input unless asked.

RUN DEEPSEEK LOCALLY USING OLLAMA AND DOCKER

To begin, make sure you set up your system properly. You can do so using this link:
https://tacticalware.com/running-ollama-on-ubuntu-24-04-with-nvidia-gpu/

Next once it is setup, you will be able to download the following models by running the command below it in bold:

DeepSeek-R1
ollama pull deepseek-r1:671b

DeepSeek-R1-Distill-Qwen-1.5B
ollama pull deepseek-r1:1.5b

DeepSeek-R1-Distill-Qwen-7B
ollama pull deepseek-r1:7b

DeepSeek-R1-Distill-Llama-8B
ollama pull deepseek-r1:8b

DeepSeek-R1-Distill-Qwen-14B
ollama pull deepseek-r1:14b

DeepSeek-R1-Distill-Qwen-32B
ollama pull deepseek-r1:32b

DeepSeek-R1-Distill-Llama-70B
ollama pull deepseek-r1:70b

Once you have downloaded and are running the model, Navigate to the web page and then select the model as shown below:

tacticalware.com and Jay Longley showing how to install DeepSeek

And that is it. You can run it locally now