Scheduler implementation of Continual Pre-Training of Large Language Models: How to (re)warm your model? #3078
Job | Run time |
---|---|
38s | |
4m 7s | |
3m 36s | |
3m 43s | |
24m 2s | |
24m 22s | |
1h 0m 28s |
Job | Run time |
---|---|
38s | |
4m 7s | |
3m 36s | |
3m 43s | |
24m 2s | |
24m 22s | |
1h 0m 28s |