-
Notifications
You must be signed in to change notification settings - Fork 264
Insights: google/maxtext
Overview
-
- 7 Merged pull requests
- 3 Open pull requests
- 0 Closed issues
- 3 New issues
Could not load contribution data
Please try again later
7 Pull requests merged by 6 people
-
Adding Mixtral-8x22b
#845 merged
Sep 5, 2024 -
Fix circ storage check for delayed case
#861 merged
Sep 5, 2024 -
Make running preflight optional in model scripts
#867 merged
Sep 5, 2024 -
documenting XLA flags used by MaxText
#848 merged
Sep 5, 2024 -
Add load balance loss
#860 merged
Sep 4, 2024 -
Add MaxText run name to TensorBoard file directory
#863 merged
Sep 4, 2024 -
Add Llama2 config for v5p
#846 merged
Sep 3, 2024
3 Pull requests opened by 3 people
-
Improve tfds perf in multihost env
#862 opened
Sep 2, 2024 -
test code to produce Lab Notes - 2024-09-07.ipynb
#866 opened
Sep 5, 2024 -
Enable expert parallelism for dropping strategy
#869 opened
Sep 6, 2024
3 Issues opened by 3 people
-
Unable to recover after checkpoint saving
#868 opened
Sep 6, 2024 -
Cannot see multiple GPUs when using Slurm (with proposed fix)
#865 opened
Sep 4, 2024 -
Converting LLama3.1 405B checkpoint - Requesting multipass checkpoint conversion
#864 opened
Sep 4, 2024
5 Unresolved conversations
Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.
-
flash attention sweep
#811 commented on
Sep 5, 2024 • 2 new comments -
mlperf gpt3 ckpt permission issues
#847 commented on
Sep 3, 2024 • 0 new comments -
Convert Orbax ckpt to HuggingFace
#581 commented on
Sep 8, 2024 • 0 new comments -
Llama3.1 (8B,70B) 🦙
#838 commented on
Sep 8, 2024 • 0 new comments -
added run_name_prefix to tensorboard
#856 commented on
Sep 3, 2024 • 0 new comments