Open LilFoxProgramer opened 1 year ago
Airoboros, a project seemingly inspired by EvolInstruct, has its dataset generation code published, but in my experience, the trained model performs significantly worse on complex instructions and factual requests. Whether its the prompt for their generation, or WizardLM's superior instruction generation method, I don't know.
Hi, I'm really curious about the algorithm behind it too. Great work! Is there a plan on releasing the code or more details about the instruction expansion algorithm?
Thank you
Hello, I hope this message finds you well. I have been following your work on the Evol-Instruct algorithm and I am very interested in learning more about it. The algorithm seems to have great potential and I believe it could be beneficial to the projects I am currently working on.
I understand that you may still be in the process of refining the algorithm, but I was wondering if you could provide any additional information on when you plan to release more details about it.