CUDA error: out of memory CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1 ...
I want to fine‑tune Gemma 3‑4B-it, and I run an evaluation after every epoch. Until 4 days ago, everything worked fine, but now I’m getting this error right ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback