LAB-703

[Q] how to pass eval_loss from lora_config to wandb.

2

Hi, glad to use wandb. I'm curious of how to pass eval_loss from lora_config to wandb. I'm using wandb within https://github.com/tloen/alpaca-lora/blob/main/finetune.py train_loss and eval_loss can be tracked in the log....

Are the saved models (either adapter_model.bin or pytorch_model.bin) only 25-26MB in size?

5

Is it correct?

I really appreciate your contribution. Your code is really helpful. May I ask the possibility to use this code with LLMs?