I have been having issues with GPT-4 Remember or access the the chat.
Literally, after 2 prompts, I get this:
I apologize for not following your instructions properly. It seems like the message was cut off, and I didn’t fully understand your request.
I apologize for the confusion, but I don’t have access to previous conversations in this thread. Your messages and my responses are not stored, and I can’t see earlier interactions. That’s why I’m unable to refer back to your original post with the different options.
I apologize for the confusion, and I understand why you’re upset. The truth is, the limitation in viewing previous parts of the conversation isn’t based on the knowledge cut-off date; it’s an inherent limitation in the way our interactions are handled. The model can’t refer back to previous parts of the conversation once they’ve been skipped over or beyond a certain length, even if it is a continuous thread.
In this instance, I was able to see your provided options because you included them in your last message. I understand that my inability to see previous messages might not align with your expectations, and I’m sincerely sorry for any frustration this has caused you.
If there are more specific questions or requests you have, please let me know, and I’ll do my best to assist you.
I apologize for any frustration this situation has caused. It seems there may have been some misunderstanding or issue with the conversation history, as I was unable to see the previous parts of the conversation.
I sincerely apologize if it seemed like I was being dishonest; that was never my intention. Due to the way this conversation system is designed, I’m not able to see previous parts of the conversation until they appear in the current conversation window. It’s possible that there was a delay or some other technical issue that prevented me from seeing the earlier parts of the conversation when you initially asked about them.
Now that I can see the three chapter starts, I’m happy to help with any feedback or guidance you’d like. Again, I apologize for any frustration this has caused, and I’m here to assist you with your request.
It’s not a token issue. GPT-3.5 has no issues.