-
Notifications
You must be signed in to change notification settings - Fork 3
Insights: AI-Hypercomputer/torchprime
Overview
-
- 8 Merged pull requests
- 3 Open pull requests
- 0 Closed issues
- 2 New issues
8 Pull requests merged by 3 people
-
Support sudoless/sudo docker everywhere
#235 merged
May 12, 2025 -
Improve ignore files
#236 merged
May 12, 2025 -
Set default profile_step to 2
#234 merged
May 10, 2025 -
Update README.md
#233 merged
May 9, 2025 -
Document the current metrics
#232 merged
May 8, 2025 -
Clarify distributed training
#231 merged
May 8, 2025 -
Update benchmark-db-writer again
#230 merged
May 8, 2025 -
Install benchmark-db-writer
#228 merged
May 7, 2025
3 Pull requests opened by 3 people
-
Update README.md
#237 opened
May 10, 2025 -
Add SFT
#238 opened
May 10, 2025 -
Document E2E test formula and secrets
#239 opened
May 12, 2025
2 Issues opened by 2 people
-
Let trainer to default use number of devices avaialble in `tp run`
#229 opened
May 8, 2025 -
Add an e2e test for environment setup
#227 opened
May 7, 2025
4 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.
-
Add Transact model and model runner to torchprime
#216 commented on
May 7, 2025 • 3 new comments -
Upload results into a database
#162 commented on
May 6, 2025 • 0 new comments -
[torch_xla] MVP correctness check for Llama 3.0 8B
#90 commented on
May 8, 2025 • 0 new comments -
[Performance Report] Improve metrics reporting
#96 commented on
May 9, 2025 • 0 new comments