Issues

• Remove tensorflow dependencies

  google/paxml#106 · Filter by author jobs-git opened on Mar 8, 2025

• Hard dependencies causing conflict with other packages

  google/paxml#105 · Filter by author jobs-git opened on Mar 8, 2025

• How to do gradient accumulation?

  google/paxml#104 · Filter by author gramesh-amd opened on Sep 27, 2024

• Question about Integration Timeline for PR #99

  google/paxml#102 · Filter by author Santiago-Castellano opened on Aug 28, 2024
  8

• LoRA support in transformers

  google/paxml#96 · Filter by author tanmayshishodia opened on Jun 27, 2024
  1

• Error running tutorial notebook

  google/paxml#95 · Filter by author ajikmr opened on Jun 21, 2024

• lingvo issue while installing paxml in vscode

  google/paxml#89 · Filter by author juneedpk opened on May 23, 2024
  1

• DEADLINE_EXCEEDED on 1024 GPUs.

  google/paxml#77 · Filter by author mhugues opened on Apr 16, 2024

• Jax + tpu and AQT int8 train model loss is abnormal

  google/paxml#71 · Filter by author Lisennlp opened on Mar 4, 2024

• Use bfloat16 for eval

  google/paxml#66 · Filter by author tbaker2 opened on Jan 26, 2024
  1

• [Question] Very low MFU(30%~35%) when train bf16 Llama2 and GPT model with single SXM4 A100 machine.

  google/paxml#65 · Filter by author MoFHeka opened on Jan 26, 2024

• [Feature Request] Need ZeRo-1/2 to cooperate with PP+TP+DP. Which may more faster than FSDP sometimes.

  google/paxml#64 · Filter by author MoFHeka opened on Jan 26, 2024

• Remaining tutorials

  google/paxml#58 · Filter by author rahulbatra85 opened on Dec 21, 2023

• How to continue training from a checkpoint?

  google/paxml#37 · Filter by author lkm2835 opened on Jul 18, 2023

• Int8 checkpoint

  google/paxml#35 · Filter by author wx-x opened on Jul 13, 2023

• ARM64 Build

  google/paxml#30 · Filter by author joker-eph opened on Jun 9, 2023
  4

• Unexpected Overheads with Activation Checkpointing with Pipeline Parallelism

  google/paxml#17 · Filter by author abhinavgoel95 opened on Apr 5, 2023
  1