@Sounds like an add-on/fix?
Yup, it’s an engineered layer on top of the core language model, so it remembers your prompts. The LLM itself is a transformer architecture, so the memory (during the training) is “preserved” via self-attention mechanism over a context window - not some persistent memory. That’s why in the past the models tended to forget old information (dropping out of the attention window). Then some tweaks were made to improve it (better preserve the most relevant information). As a fun fact, it seems closer to the brain memory model of raisins in the cake than eg LSTM (no persistent memory cells, but selectively attended and propagated information- kind of emergent).
I hope this makes sense- just climbed a big mountain, pleasantly weary.
Serving the Quantitative Finance Community
-
- katastrofa
- Posts: 7929
- Joined:
- Location: Event Horizon
Re: Using AI for software development
Last edited by katastrofa on July 20th, 2025, 6:46 am, edited 1 time in total.
-
- Cuchulainn
- Posts: 22924
- Joined:
Re: Using AI for software development
ChatGPT OpenAI Product development meeting?It does have memory: https://openai.com/index/memory-and-new ... r-chatgpt/
-
- Cuchulainn
- Posts: 22924
- Joined:
Re: Using AI for software development
Memory?? it just saves a history list in a dictionary, methinks. Ebbinghaus would roll in his grave.It does have memory: https://openai.com/index/memory-and-new ... r-chatgpt/
The Pro version costs $229 per month but has RBAC (access control, Pandora's box). Too difficult for 90% of software orgs.
I think it is a pre-sales kite-flying attempt at railing in gullible clients. Doesn't grow on trees.
Last edited by Cuchulainn on July 20th, 2025, 12:39 pm, edited 1 time in total.
-
- Cuchulainn
- Posts: 22924
- Joined:
Re: Using AI for software development
That's pretty clear; takk.@Sounds like an add-on/fix?
Yup, it’s an engineered layer on top of the core language model, so it remembers your prompts. The LLM itself is a transformer architecture, so the memory (during the training) is “preserved” via self-attention mechanism over a context window - not some persistent memory. That’s why in the past the models tended to forget old information (dropping out of the attention window). Then some tweaks were made to improve it (better preserve the most relevant information). As a fun fact, it seems closer to the brain memory model of raisins in the cake than eg LSTM (no persistent memory cells, but selectively attended and propagated information- kind of emergent).
I hope this makes sense- just climbed a big mountain, pleasantly weary.
-
- Cuchulainn
- Posts: 22924
- Joined:
Re: Using AI for software development
CSV?NB today I was trying to figure out how to parse an unknown data format at work, and I just copied a sample into Copilot, asked it "how do I decode it?" and it gave me a working answer.
Binary or ASCII/EBCDIC?
-
- Cuchulainn
- Posts: 22924
- Joined:
Re: Using AI for software development
And you need AI to do that??Believe it or not, it's called SWIFT.
-
- Cuchulainn
- Posts: 22924
- Joined:
Re: Using AI for software development
You mean that what?To do what?