14
votes
Accepted
Can an LLM be constrained to answer questions only about a specific dataset?
There are a few approaches you could take to constrain an AI to only answer questions based on a specific dataset:
Fine-tuning: Take an existing large language model like GPT-3 and fine-tune it on ...
- 526
6
votes
Can an LLM be constrained to answer questions only about a specific dataset?
You can't restrict an LLM's answers either through prompting or through training. Restrictions in the prompt can be overridden by the asker being sufficiently clever in wording the question (for ...
- 1,191
4
votes
Is it possible to create a reliable AI quotation tool using excel spreadsheets with ChatGPT API
For example, if a company hasn't had a particular task done in 4 years and we've had 3 price increases during that time, the AI should calculate the correct amount for that company considering ...
- 2,099
4
votes
Accepted
ChatGPT API Embedding Vectors - Are vectors standard across LLM's?
Embeddings are not universal, they vary per framework/API.
Read an intro here: https://gpt-index.readthedocs.io/en/latest/core_modules/model_modules/embeddings/root.html
And then see various ...
- 384
3
votes
Which techstack and IDE are good to set up a RAG model? (=Retrieval-augmented generation models jointly fine-tune DPR and sequence-to-sequence models)
Regarding RAG:
Python + the LangChain lib for the retriever + GPT for the answer generator is a common choice, but there exist many alternatives.
Code example for RAG-QA to answer questions about how ...
- 4,181
2
votes
Can an LLM be constrained to answer questions only about a specific dataset?
Prashanth, CEO's Stack Overflow Inc., this week announced OverflowAI with promises to deliver content based on Stack Overflow. Considering this and how much you might trust Stack Overflow as a ...
- 1,025
2
votes
Is there any AI apis that can remember your information?
there are a few ways to handle the conversational history, but one mainstream method is that you as the application developer handle this. for example, you send to the LLM api the following (the part ...
- 121
2
votes
Accepted
How much RAM is required for model inferencing?
GGUF = new file format aiming at replacing GGML.
Yes.
(V)RAM requirement for inference: Model size < RAM requirements < Model size * 1.2.
https://huggingface.co/spaces/hf-accelerate/model-...
- 4,181
1
vote
Accepted
What is model inferencing - computation or look-up?
Yes. 3. Stateless between two subsequent inferences. 4. Add or remove model instances eg via Kubernetes depending on API usage.
- 4,181
1
vote
Is there any AI apis that can remember your information?
I want to augment to @NirO.'s answer. Yes, it is theoretically possible to recall the user's identity from the chat-history. But there are some practical issues here like:
When the chat becomes too ...
- 171
1
vote
Accepted
What's the difference between "Tokens per Minute Rate Limit (thousands)" and "Rate limit (Tokens per minute)" in Azure OpenAI?
If I edit "Tokens per Minute Rate Limit (thousands)" in the deployment of the model e.g. to 27k tokens, it'll change Tokens per Minute Rate Limit (thousand) to 27 and Rate limit (Tokens per ...
- 4,181
1
vote
chatGPT messages not always in French, sometimes in English
I've had this happen when using GPT-3 (at the time) as part of a language learning app. It seemed to pick up on the fact that the inputs were not from a native speaker, and the more noticeable that ...
- 161
1
vote
How can I get chatgpt to keep track of all the words in the list that I am providing?
The problem of ChatGPT is that it seems to guess its statistics, it does not run a program for it. At least I tried to get the easy task done that it would just tell me the running total number of ...
- 183
1
vote
How can I get chatgpt to keep track of all the words in the list that I am providing?
How can I fix this?
We'd need more examples to try to guess what's going on (e.g., maybe the issue in this specific example is that "metal" only appears in "heavy metal")
One ...
- 4,181
1
vote
How do I force ChatGPT API response to be brief while being brief myself?
Just like @Wicket suggested, I have naturally been rewording the prompt a lot, but it wasn't getting any better, at least not much.
No matter how I worded my prompt, the response would still always ...
- 163
1
vote
Accepted
How do I force ChatGPT API response to be brief while being brief myself?
If a wording didn't work as you expect try changing it. In this case the system prompt used was
Briefly point out grammatical errors in this sentence
Try rewording the about using a direct and more ...
- 1,025
1
vote
ChatGPT API Embedding Vectors - Are vectors standard across LLM's?
There is no standardization for embedding vectors across different Large Language Models (LLMs). Each LLM trains its own embeddings as part of the model architecture and training process. As a result, ...
- 526
Only top scored, non community-wiki answers of a minimum length are eligible