Discussions

• 3 You must be logged in to vote

  🙏

  How to enable context parallelism?

  alex-ht asked Apr 11, 2024 in Q&A · Answered

  

  

  

  

  13
• 1 You must be logged in to vote

  🙏

  load_gbs in build_dataloader, how it works?

  shensimeteor asked Dec 17, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  Does Nemo-Aligner supports gradient accumulation (accumulate_grad_batches)?

  shensimeteor asked Dec 14, 2024 in Q&A · Unanswered

  

  

  2
• 1 You must be logged in to vote

  🙏

  [QUESTION] get different reward scores for the same prompt

  Cppowboy asked Dec 10, 2024 in Q&A · Unanswered

  

  

  0
• 2 You must be logged in to vote

  🙏

  Context Parallelism in train_gpt_dpo.py

  Wolfwjs asked Aug 26, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  Unable to start reward-model-and-critic-server

  pratikkumar018 asked Feb 20, 2024 in Q&A · Unanswered

  

  

  

  

  19
• 📣

  Welcome to NeMo-Aligner Discussions!

  okuchaiev announced Jan 5, 2024 in Announcements

  

  

  0
• 1 You must be logged in to vote

  🙏

  Is there a limit on the largest PPO model size and max context length the package can support to scale out of box? Is it scalable when we add more GPU nodes to the training job? Thanks! Sent from my iPhoneOn Jan 17, 2024, at 11:12 AM, HeyyyyyyG ***@***.***> wrote:

  okuchaiev asked Jan 22, 2024 in Q&A · Unanswered

  

  

  1
• 1 You must be logged in to vote

  🙏

  What's the largest PPO model size and context length that have been trained successfully with this library? Can you also share some performance metrics (i.e. GPU count, training time) if possible?

  okuchaiev asked Jan 22, 2024 in Q&A · Unanswered

  

  

  1