Does this model work?
#1
by
PSM272
- opened
Hello, I saw this model from your dataset, and, I was wondering if it worked (or well, at least)...
The chat template is really finicky right now (the model never quits generating, I think I made a stupid mistake somewhere in the training code) and I'm retraining/re-fine-tuning a new one with ChatML template which will hopefully be done training today!
I'll let you know when I get it working :)
Edit: Oh and also, the v0.0 and v0.1 versions aren't working right now...
Update: The new model's up (it's correct this time)!