Hi, I am new to this GPUGrid project and encountering a strange error(195 (0xc3) EXIT_CHILD_FAILED) when Boinc tries to run this specific task called ATMML: Free energy with neural networks v1.01 (cuda1121)
I checked the task logs and I guess the error is:

Traceback (most recent call last):
  File "E:\BOINC\slots\11\Scripts\rbfe_explicit_sync.py", line 11, in <module>
    rx.scheduleJobs()
  File "E:\BOINC\slots\11\Lib\site-packages\sync\atm.py", line 126, in scheduleJobs
    self.worker.run(replica)
  File "E:\BOINC\slots\11\Lib\site-packages\sync\worker.py", line 124, in run
    raise RuntimeError(f"Simulation failed {ntry} times!")
RuntimeError: Simulation failed 5 times!

I don't understand anything in this log. I haven't overclocked my GPU. I wonder if my GPU does not support GPUGrid
You can check the task log here: https://www.gpugrid.net/result.php?resultid=36661000
a Pastebin link if the link above does not work: https://pastebin.com/Ty2gkUHV

I would greatly appreciate any help. Thank you!

Information about my computer:
Boinc version: 8.0.2
BOINC computer preferences: https://imgur.com/a/IEEj2Fl
Operating system: Windows 10 (64x)
CPU: AMD FX(tm)-8150 Eight-Core Processor [Family 21 Model 1 Stepping 2] (CPU is overclocked)
GPUS(I have 2 GPUs): NVIDIA GeForce GTX 970 (4095MB) driver: 566.14, NVIDIA GeForce GTX 750 Ti (2048MB) driver: 566.14
CUDA software version(I decided to install the latest CUDA runtime from the official Nvidia page but no luck): 12.6(latest)
DISK STORAGE that I'm using for BOINC: 1TB USB HDD

Questions like this are best asked on the message boards of the specific project impacted. I answered this particular question yesterday, in https://www.gpugrid.net/forum_forum.php?id=1.

Ah, okay. Thank you for the information.