mirror of https://github.com/InternLM/InternLM
* catch exception of all ranks * monitor task only if DO_ALERT is True |
||
|---|---|---|
| .. | ||
| legacy | ||
| __init__.py | ||
| initialize_tensor.py | ||
| initialize_trainer.py | ||
| launch.py | ||
* catch exception of all ranks * monitor task only if DO_ALERT is True |
||
|---|---|---|
| .. | ||
| legacy | ||
| __init__.py | ||
| initialize_tensor.py | ||
| initialize_trainer.py | ||
| launch.py | ||