Closed abhinavg4 closed 1 month ago
Llama-style experiment mixture support.
Apart from this, you need to provide initialize_from_checkpoint_path in the config and also change the data config to 0.7 and 0.3 weights.
initialize_from_checkpoint_path
This looks good. do you have a wandb run I can see?
Run looks fine. going to merge.
https://wandb.ai/stanford-mercury/marin/runs/llama1b-fw-txt-dclm-mixture-0825?nw=nwuserabhinavg4
Llama-style experiment mixture support.
Apart from this, you need to provide
initialize_from_checkpoint_path
in the config and also change the data config to 0.7 and 0.3 weights.