Skip to content

Add CUDA implementation for attn_probs

239df8b
Select commit
Loading
Failed to load commit list.
Closed

Make MultiHeadAttention op return attention probabilities #23125

Add CUDA implementation for attn_probs
239df8b
Select commit
Loading
Failed to load commit list.
This check has been archived and is scheduled for deletion. Learn more about checks retention
Azure Pipelines / Linux DNNL CI Pipeline succeeded Dec 17, 2024 in 1h 19m 7s

Build #20241217.1 succeeded

Details

Tests

  • Failed: 0 (0.00%)
  • Passed: 10,302 (99.93%)
  • Other: 7 (0.07%)
  • Total: 10,309