Redo custom attention processor to support other attention types #2296
Job | Run time |
---|---|
4m 45s | |
5m 29s | |
4m 30s | |
4m 45s | |
8m 28s | |
4m 40s | |
4m 33s | |
3m 28s | |
5m 20s | |
15m 3s | |
1h 1m 1s |
Job | Run time |
---|---|
4m 45s | |
5m 29s | |
4m 30s | |
4m 45s | |
8m 28s | |
4m 40s | |
4m 33s | |
3m 28s | |
5m 20s | |
15m 3s | |
1h 1m 1s |