ItsLogic

> @ItsLogic Guys, do you remember what was the minimal loss when you stopped the training? I hovered around low 0.8 and high 0.7 from about 20% through the first...

> Any hero out there got Alpaca-13B distributed yet? For those of us that lack a 24GB GPU sweat I uploaded my epoch 1 and epoch 3 loras https://huggingface.co/Draff/llama-alpaca-stuff/tree/main/Alpaca-Loras

> @ItsLogic Forgive my ignorance, but how do I download this? `from_pretrained()` doesn't support subdirectories. just download [adapter_config.json](https://huggingface.co/Draff/llama-alpaca-stuff/blob/main/Alpaca-Loras/13b-e3/adapter_config.json) and [adapter_model.bin](https://huggingface.co/Draff/llama-alpaca-stuff/blob/main/Alpaca-Loras/13b-e3/adapter_model.bin) manually and put them in a folder and then edit...

> Any noticeable difference between epoch 1 and epoch 3? Havent done any in depth testing yet but from my usage so far they feel about the same

event based games are now supported there are no plans for 360

> I understand that but the whole point is to get help to figure out what the issue is...... I can launch Version 2.2 and log in with no issues....

Thanks for this and sorry for just not responding for 5 months (oops) I am going to write some yml issue templates though

This is sub optimal. Ill have a look into it tomorrow to see if I can reproduce

closing this as filters should now be fixed in latest version. Reopen if the bug still happens