Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

About augmentation with Change specific numbers. #9

Open
ChengpengLi1003 opened this issue Mar 26, 2024 · 1 comment
Open

About augmentation with Change specific numbers. #9

ChengpengLi1003 opened this issue Mar 26, 2024 · 1 comment
Labels
wontfix This will not be worked on

Comments

@ChengpengLi1003
Copy link

ChengpengLi1003 commented Mar 26, 2024

when I use "Change specific numbers:" to augment problem using internlm2-math-20b to, the model answering the question instead of augment it~

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("internLLMATH", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("internLLMATH", trust_remote_code=True, torch_dtype=torch.float16).cuda()
model = model.eval()
response, history = model.chat(tokenizer, "Change specific numbers: Find the center of the circle with equation $x^2 - 6x + y^2 + 2y = 9$", history=[], meta_instruction="")
print(response)

We complete the square for both $x$ and $y$ terms. For the $x$ terms, we add $(6/2)^2 = 9$ to both sides. For the $y$ terms, we add $(2/2)^2 = 1$ to both sides. This gives us:
$(x^2 - 6x + 9) + (y^2 + 2y + 1) = 9 + 9 + 1$
$(x - 3)^2 + (y + 1)^2 = 19$
So, the center of the circle is $\boxed{(3, -1)}$.
The answer is: (3,-1).

@objecti0n
Copy link
Collaborator

Data augmentation data is not too much in SFT data, so this is possible. Please consider sample multiple times.

@objecti0n objecti0n added the wontfix This will not be worked on label Jun 27, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
wontfix This will not be worked on
Projects
None yet
Development

No branches or pull requests

2 participants