What is the estimated time required to fine-tune the Llama 2 model?

Question

What is the estimated time required to fine-tune the Llama 2 model?

Chen, Xiaoyi 0

I have fine-tuned the Llama-2-7B model on my customized dataset of just 50 samples, setting it to run for 2 epochs. However, it has been running for over a day (shown as the screenshot), significantly exceeding the usual 4-hour completion time for Llama models. I am concerned about the extended processing time. Could this be due to a queue, or is there a premium service available to expedite the process? This is an urgent matter for me, so I would appreciate your prompt assistance. llama

1 answer

Your answer

Answer 1

Ramr-msft 17,826

Thanks for the question, Can you please add more details about the VM that you have selected. Please check the quota page. User's image

Chen, Xiaoyi 0 Reputation points

2024-02-21T21:41:55.09+00:00

I didn't find any pages that can check this VM setting. My setting page is shown as follows. I wonder if there is something wrong.
Ramr-msft 17,826 Reputation points

2024-03-04T04:42:28.2433333+00:00

Thanks for the details, you check the qutoa in the following link and available by region.

Share via

What is the estimated time required to fine-tune the Llama 2 model?

1 answer

Your answer