iamkylebalmer
Region: GB
Wednesday 31 July 2024 11:01:48 GMT
20000
682
61
110
Music
Download
Comments
Hagar Listens | מוזיקה ישראלית :
Interesting! Thanks for the insights. Does that also apply to Custom GPT?
2024-07-31 21:11:00
1
Citronic :
This makes perfect sense based on my experience. Thank you for explaining it. Do you think Claude is better than Chat GPT? Or should I pay for both of them? Thank you
2024-07-31 18:42:41
9
Natural_Nat_Tech :
Almost like a goldfish memory 😅
2024-07-31 11:16:07
7
nn :
OK but what's the work around now I know?
2024-07-31 21:50:37
2
Reksaurian :
While that's interesting, I find its best to treat it like a single QnA. You ask once, done, new chat, prompt with the information you just generated or a Q about what you need specifically right now.
2024-07-31 12:37:44
5
Margus Martsepp :
As an expert - not quite. LLM doesn't know it's own context window size, nor can it tell you accurate info - it is guess-estimating, based on previous models info that was part of it's training data.
2024-10-23 14:50:49
2
Kirsty :
so would it be better to create a GPTs with a knowledge base to work with?
2024-07-31 18:16:08
1
Vokt AI :
Claude is fantastic regarding memory and there new chain prompting is a rag killer
2024-10-06 00:04:14
0
Jack :
I wonder if there's a pre amble that could be added to a prompt to make use of this. Maybe, check your context window, if you have exceeded it, tell me, if not, continue. Could be good for CustomGPTs
2024-07-31 12:13:36
2
Fiona in the Wild :
Very useful (also I’m going to apply this to to kids and ask them ‘how much of what I’ve already told you to do are you remembering at this point?’ 😅)
2024-07-31 13:12:25
3
Peter Mesarec | AI v podjetjih :
If you use the paid version, the context windows is much higher. 128.000 tokens.
2024-07-31 11:18:20
2
Moe Jason :
If working with a larger document or set of info, can I also ask chat gpt to e.g. identify/list the key information in as few words as possible. Then use the response in another chat to reduce the number of tokens used?
2024-07-31 11:10:33
5
🈲 :
I tried but nothing came out. The answer was different totally different
2024-07-31 13:50:17
2
Gharibat :
I understand now the systematic mess
2025-08-23 11:24:13
1
TechIfat :
I used to get frustrated after overwhelming chatgpt with large files lol thanks for pointing out the reason behind its dementia 😵💫
2024-08-22 14:49:53
0
Maya :
What about when I ask chatgpt to summarize the whole website, hiw does the memory wotk then related to processed words?
2024-10-21 05:24:31
1
Axel Windbrake :
Would it help to use larger files in the vector store and use an assistant to overcome this context window issue? Do you know if ChatGPT uses embeddings in the vector store documents
2024-08-02 14:19:52
1
dizzymoore1 :
This validates what I have been experiencing and not being able to understand
2024-08-12 10:30:37
1
Sunshine 🌞 :
wow thanks
2024-07-31 11:53:20
2
Paddy :
Yes I’ve had it. I caught one mistake but didn’t the other. It made up a quote and I didn’t catch it. Caused some problems
2024-08-26 19:50:34
1
Blessed and Busy 🫶🏼❤️ :
It has 8000 tokens per every new chat?
2024-07-31 13:29:44
2
King.3amri :
So what’s the solution here ? Create a new chat conversation whenever i reach the limit ? And how to reference to another chat ?
2024-08-01 07:42:36
1
DPC :
My luck, asking about the context window. Will immediate lead to tokens spent 🤣
2024-08-03 22:20:59
1
Leon Yohai :
Listen and explain me: I’m paying 200$/mo on ChatGPT. And it has only 4,000 context window… the last 3 days can’t do anything as u said… open AI says all ok… I’m quitting
2025-05-19 16:35:29
1
Godwar 🪽 🪽 🪽 :
🔥🔥 🔥i didn't know how to check it. useful tiktok
2024-07-31 19:21:22
1
To see more videos from user @iamkylebalmer, please go to the Tikwm
homepage.
