Michael
Michael
# Before submitting - [x] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [x] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [x] Did you...
## 🐛 Bug If you train with multigpu using ddp while logging with `wandb`, then the program doesn't exit as wandb hangs because there is no `wandb.finish()` call. See https://docs.wandb.ai/guides/track/advanced/distributed-training#hanging-at-the-end-of-training...
Printing extension had default print_status=True without the ability to change it, I'm just adding it as a an argument. Also added tests to a new `test_printing.py`, I found there was...
addresses #1567 changed name of parameter from `revision` to `base_model_revision` for clarity can add unit tests if someone gives pointers to similar ones to extend may need to add a...
### Feature request Currently, the `adapter_config` only stores the `base_model_name_or_path` but not the `revision`. This means that if a `PeftModel` is trained on top of a model with a `revision`...
In running experiments on IMDB, I found that there was a very high variance in validation and test set results and I don't fully understand it, so I'm looking for...
Hi, I'm currently running the imdb experiments and trying to reproduce the PPO and NLPO results from the paper and though my PPO is close, NLPO is quite far from...