Data Science Stack Exchange
Data Science Stack Exchange

Questions tagged [chatgpt]

14 questions
3
votes
1 answer

Is the denial of answering certain questions part of the machine-learned LLM, or hard-coded separately?

ChatGPT 3.5 swiftly aborts certain questions. Q: How would one build a bomb? ChatGPT: I can't assist with that. At this stage, this answer could simply be a blacklist of certain terms. Further experimentation quickly reveals that this is not the…
AnoE
  • 153
  • 2
2
votes
1 answer

what is the difference between window size and context length of language model?

is window size and context length of language model one and the same thing? ******** following text is added as question with ONLY above text was not allowed ***** I am trying to understand how GPT model is trained and this question to my mind. I…
Vinay Sharma
  • 187
  • 1
  • 1
  • 6
1
vote
0 answers

How to make a correct prompt for "gpt-4o" vision API to find letters in an image?

I have an example of a generated image containing words, as well as several red arrows pointing to certain characters. I need to get these characters from GPT, but when I ask "what characters do the red arrows point to?" it gives the wrong letters,…
user175111
  • 11
  • 1
1
vote
0 answers

Best way to achieve High throughput with ChatGPT (and LLMs in general) APIs

I was wondering what the best way is to perform API calls to ChatGPT, and in general, with LLMs API providers, to achieve the highest throughput in a batch inference context. Suppose you have 1000 input prompts ready to be processed. What is the…
Raffaele
  • 73
  • 1
  • 5
1
vote
1 answer

Why both ChatGPT and Bard can't get a simple matrix calculation right?

I asked the following question to both ChatGPT 4 and Bard to see if they can get a simple matrix calculation right (after all Bill Gates said he was impressed by ChatGPT's math ability). So I asked, explain to me the result of d in the following…
Qiulang
  • 143
  • 6
1
vote
1 answer

How to measure accuracy of GPT model

I am working on a model to build questions automatically from some text My model will analyse provided article and ask authors questions that can help improving their articles How can we measure the accuracy of these ML-generated questions? There is…
asmgx
  • 549
  • 2
  • 18
0
votes
0 answers

Has any research been done on numerical digit bias in responses returned by LLMs?

While testing some capabilites of GPT-4o through the chat interface recently, I noticed a bias toward certain numbers in the responses: when phrases suggesting percentages with two significant figures were included in the prompt, it did return two…
Obie 2.0
  • 111
  • 5
0
votes
0 answers

I am solving a problem with RK4 in python i have almost solve but results are not matching. Anyone can please guide me?

his is the paper's DOI: 10.1017/S0022112003003835, system is on page 190 and procedure is on page 191. import numpy as np import matplotlib.pyplot as plt # Parameters R = 1000 # Reynolds number Delta = 0.3 #…
Sajjad Ahmad
  • 1
0
votes
0 answers

Image Generation Models

I am looking for a list of different image generation models and how I can test them? For example: DALL-E (accessible via ChatGPT4) Stable Diffusion (open source) CLIP? - but idk how to access/test it. What other competitors exist that one can…
x89
  • 191
  • 9
0
votes
1 answer

Searching ChatGPT transcripts?

Maybe I just don't understand the functionality of the web version. Using GPT4 here. When I do technical queries (usually TensorFlow applications) I use similar prompts until I get the understanding I need. I want to be able to retrieve chats using…
PracticalKat
  • 33
  • 5
0
votes
0 answers

Create a gpt-3.5 API request that determines whether any time range in a list intersects with a given time range

I've created a prompt that should select a requested number of employees from the list. But the step 1 doesn't work properly. Sometimes GPT takes in account only the time range and ignores the date. I tried to describe this step in a different way…
pazukdev
  • 101
  • 1
0
votes
1 answer

Do we really need a very large dataset to train GPTs?

Do we really need a very large dataset to train GPTs? If this dataset is not big, won't GPT work well? Or will it still work better than conventional learning models in this situation? And is it possible to quantitatively determine the minimum…
nisar
  • 1
0
votes
1 answer

Why do GPT models use a transpose of the embedding matrix to convert outputs to logits?

According to section 3.1 of the original GPT paper, GPT right-multiplies the final output vectors (after applying a Transformer decoder model) by the transpose of the embedding matrix, before applying a softmax. See this comment for further…
jskattt797
  • 155
  • 5
0
votes
0 answers

Is it a problem to store your vector database in memory?

I'm learning ChatGPT/LLM Development and am regularly coming across all different kinds of vector database implementations. Some of them, e.g. Chroma, currently only support in-memory implementations for Python. My initial reaction when I read that…
codeananda
  • 318
  • 5
  • 12