Normalize gates on expert dim before calculating seq_aux_loss #21889
Triggered via pull request
November 3, 2025 12:37
lshpku
Status
Cancelled
Total duration
1d 0h 0m 2s
Artifacts
–
Annotations
1 error
|
Lint
The job has exceeded the maximum execution time while awaiting a runner for 24h0m0s
|