I am confused and i’m wondering if anyone else is experiencing what i have been going though. For the past 3 months i have been attempting to find tune a model on my own computer using HF Autotrain-Advanced, 3 different times the program seemed to be working like clockwork and then all of a sudden…NOTHING! What i mean by that is that the program would stop and there would be no trace of the fine tuned model on my computer, my HF account or anywhere. I have checked the HF folder within the .cache file, i have checked the conda env the fine tuning was being done in also and nothing. So i attempted to give it one last try before going to another platform like Together Ai to try and get this job done but today when i logged into HF it is showing that i am Autotraining a model on there platform when i am not! I am doing it on my own computer that has 2 NVIDIA GeForce RTX 4070 SUPER 12GB GDDR6X and the processor is a intel i9 which is why i attempted to fine tune locally in the first place. It seems that the site is tracking the fine tuning process while its going on locally which is a problem because what happed to the models on the other 2 attempt i tried? Documentation for those models are on my account but no LLM. I opted out of the push to hub option as well so no documentation or anything else should be on my account right? Or is all this normal? I feel like i have used my own resources to fine tune 2 models for someone else. Can someone please explain whats going on? Ine More thing…VsCode is showing 1, processor that isn’t the ones i have on my PC and the conda env was created within a docker container if that matters in providing guidance.
